paperID
stringlengths
36
36
pwc_id
stringlengths
8
47
arxiv_id
stringlengths
6
16
nips_id
float64
url_abs
stringlengths
18
329
url_pdf
stringlengths
18
742
title
stringlengths
8
325
abstract
stringlengths
1
7.27k
authors
stringlengths
2
7.06k
published
stringlengths
10
10
conference
stringlengths
12
47
conference_url_abs
stringlengths
16
198
conference_url_pdf
stringlengths
27
199
proceeding
stringlengths
6
47
taskID
stringlengths
7
1.44k
areaID
stringclasses
688 values
embedding
stringlengths
9.26k
12.5k
umap_embedding
stringlengths
29
44
ae1893c0-3787-4b35-984d-8b681b49236f
sequential-local-learning-for-latent
1703.04082
null
http://arxiv.org/abs/1703.04082v2
http://arxiv.org/pdf/1703.04082v2.pdf
Sequential Local Learning for Latent Graphical Models
Learning parameters of latent graphical models (GM) is inherently much harder than that of no-latent ones since the latent variables make the corresponding log-likelihood non-concave. Nevertheless, expectation-maximization schemes are popularly used in practice, but they are typically stuck in local optima. In the rece...
['Sejun Park', 'Eunho Yang', 'Jinwoo Shin']
2017-03-12
null
null
null
null
['novel-concepts']
['reasoning']
[ 1.73684031e-01 2.85867691e-01 -3.51872742e-01 -2.19481558e-01 -8.12169969e-01 -3.50668997e-01 5.63534021e-01 -2.20496237e-01 -2.06145421e-01 7.35459983e-01 5.67432791e-02 -4.11384016e-01 -4.31235209e-02 -7.99792886e-01 -7.02174485e-01 -1.11378145e+00 -1.00821711e-01 4.22511607e-01 3.12946066e-02 2.57501364...
[7.159030914306641, 3.964144229888916]
cc265cf1-63dc-432c-acdc-d9bf85996223
learning-representations-of-bi-level
2302.02601
null
https://arxiv.org/abs/2302.02601v3
https://arxiv.org/pdf/2302.02601v3.pdf
Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction
Knowledge graphs represent known facts using triplets. While existing knowledge graph embedding methods only consider the connections between entities, we propose considering the relationships between triplets. For example, let us consider two triplets $T_1$ and $T_2$ where $T_1$ is (Academy_Awards, Nominates, Avatar) ...
['Joyce Jiyoung Whang', 'Chanyoung Chung']
2023-02-06
null
null
null
null
['knowledge-graph-embedding']
['graphs']
[-2.87644833e-01 5.00832915e-01 -3.92342925e-01 -3.32445920e-01 -3.77704352e-01 -4.45982635e-01 3.45020592e-01 5.50128341e-01 -1.22671001e-01 1.00250256e+00 -1.59760807e-02 -5.28083205e-01 -6.93050086e-01 -1.48512101e+00 -1.13297009e+00 -1.99402645e-01 -4.85479146e-01 6.77703917e-01 2.58500844e-01 -3.81476521...
[8.774564743041992, 7.837462425231934]
0f2b50b3-7a55-41fd-9b12-cde452def348
predicting-ground-level-scene-layout-from
1612.02709
null
http://arxiv.org/abs/1612.02709v1
http://arxiv.org/pdf/1612.02709v1.pdf
Predicting Ground-Level Scene Layout from Aerial Imagery
We introduce a novel strategy for learning to extract semantically meaningful features from aerial imagery. Instead of manually labeling the aerial imagery, we propose to predict (noisy) semantic features automatically extracted from co-located ground imagery. Our network architecture takes an aerial image as input, ex...
['Zachary Bessinger', 'Menghua Zhai', 'Scott Workman', 'Nathan Jacobs']
2016-12-08
predicting-ground-level-scene-layout-from-1
http://openaccess.thecvf.com/content_cvpr_2017/html/Zhai_Predicting_Ground-Level_Scene_CVPR_2017_paper.html
http://openaccess.thecvf.com/content_cvpr_2017/papers/Zhai_Predicting_Ground-Level_Scene_CVPR_2017_paper.pdf
cvpr-2017-7
['cross-view-image-to-image-translation']
['computer-vision']
[ 6.74173534e-01 5.95186949e-01 1.79711923e-01 -6.51662290e-01 -7.62790203e-01 -1.03144419e+00 4.39456999e-01 1.52368829e-01 -4.08839017e-01 2.34680459e-01 6.60612732e-02 -1.73245013e-01 4.11084332e-02 -1.13746929e+00 -9.88372028e-01 -2.75667995e-01 -2.39502221e-01 2.91767538e-01 8.47833753e-02 -1.42232165...
[9.405879974365234, -0.16569675505161285]
fd56a1ad-5858-4477-9abb-2647391b842b
positive-unlabeled-learning-for-binary-and
2302.08050
null
https://arxiv.org/abs/2302.08050v1
https://arxiv.org/pdf/2302.08050v1.pdf
Positive-unlabeled learning for binary and multi-class cell detection in histopathology images with incomplete annotations
Cell detection in histopathology images is of great interest to clinical practice and research, and convolutional neural networks (CNNs) have achieved remarkable cell detection results. Typically, to train CNN-based cell detection models, every positive instance in the training images needs to be annotated, and instanc...
['Chuyang Ye', 'Zhiwen Liu', 'Yaou Liu', 'Fengqian Pang', 'Zipei Zhao']
2023-02-16
null
null
null
null
['cell-detection']
['computer-vision']
[ 3.38561624e-01 7.45252073e-02 -3.99027467e-01 -2.44490609e-01 -6.72533333e-01 -4.95289207e-01 6.70510307e-02 7.17106223e-01 -7.77433753e-01 1.14036930e+00 -5.23315012e-01 -3.21835607e-01 3.62622589e-01 -9.51052487e-01 -4.37582701e-01 -1.25054729e+00 8.72982666e-02 5.68862379e-01 3.82313848e-01 2.79184312...
[14.900774002075195, -3.0699961185455322]
8b482749-060e-454b-9f04-859bce827b00
paraphrase-generation-as-unsupervised-machine
2109.02950
null
https://arxiv.org/abs/2109.02950v2
https://arxiv.org/pdf/2109.02950v2.pdf
Paraphrase Generation as Unsupervised Machine Translation
In this paper, we propose a new paradigm for paraphrase generation by treating the task as unsupervised machine translation (UMT) based on the assumption that there must be pairs of sentences expressing the same meaning in a large-scale unlabeled monolingual corpus. The proposed paradigm first splits a large unlabeled ...
['Fei Wu', 'Xiaofei Sun', 'Jiwei Li', 'Nanyun Peng', 'Yuxian Meng', 'Yufei Tian', 'Chun Fan']
2021-09-07
null
https://aclanthology.org/2022.coling-1.555
https://aclanthology.org/2022.coling-1.555.pdf
coling-2022-10
['paraphrase-generation', 'unsupervised-machine-translation', 'paraphrase-generation']
['computer-code', 'natural-language-processing', 'natural-language-processing']
[ 4.16303605e-01 6.01404021e-03 -3.27606052e-01 -6.27356768e-01 -9.45417166e-01 -8.34177256e-01 6.72004998e-01 -1.18788883e-01 -1.33883715e-01 1.01589954e+00 3.17052692e-01 -4.83477116e-01 3.34087372e-01 -6.34437442e-01 -7.00571120e-01 -6.31770730e-01 8.92423093e-01 7.90714145e-01 -2.10596919e-01 -2.70540804...
[11.663334846496582, 9.431385040283203]
b870fd70-5655-4f29-813c-278c03296456
causes-of-catastrophic-forgetting-in-class
2209.08010
null
https://arxiv.org/abs/2209.08010v1
https://arxiv.org/pdf/2209.08010v1.pdf
Causes of Catastrophic Forgetting in Class-Incremental Semantic Segmentation
Class-incremental learning for semantic segmentation (CiSS) is presently a highly researched field which aims at updating a semantic segmentation model by sequentially learning new semantic classes. A major challenge in CiSS is overcoming the effects of catastrophic forgetting, which describes the sudden drop of accura...
['Jürgen Beyerer', 'Tobias Kalb']
2022-09-16
null
null
null
null
['class-incremental-semantic-segmentation']
['computer-vision']
[ 7.14708686e-01 4.94656920e-01 1.93332620e-02 -3.71116340e-01 -3.45404744e-01 -4.05616403e-01 5.55549145e-01 3.90361249e-01 -6.32904649e-01 8.90174389e-01 1.09960176e-01 1.37475625e-01 1.04726866e-01 -7.21387029e-01 -1.11775672e+00 -8.05175304e-01 3.19602787e-01 4.40300912e-01 7.30597079e-01 -5.54129593...
[9.37452220916748, 2.419447660446167]
8873c68f-3a2a-47d1-be36-52768b2da034
using-simulation-and-domain-adaptation-to
1709.07857
null
http://arxiv.org/abs/1709.07857v2
http://arxiv.org/pdf/1709.07857v2.pdf
Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
Instrumenting and collecting annotated visual grasping datasets to train modern machine learning algorithms can be extremely time-consuming and expensive. An appealing alternative is to use off-the-shelf simulators to render synthetic data for which ground-truth annotations are generated automatically. Unfortunately, m...
['Laura Downs', 'Yunfei Bai', 'Matthew Kelcey', 'Vincent Vanhoucke', 'Mrinal Kalakrishnan', 'Alex Irpan', 'Sergey Levine', 'Paul Wohlhart', 'Peter Pastor', 'Kurt Konolige', 'Konstantinos Bousmalis', 'Julian Ibarz']
2017-09-22
null
null
null
null
['industrial-robots']
['robots']
[ 4.86306965e-01 1.42951692e-02 2.91985035e-01 -4.47026938e-01 -7.60884404e-01 -9.83786702e-01 4.21908081e-01 -1.01210497e-01 -5.30580044e-01 9.15142715e-01 -5.49345076e-01 -6.21158350e-03 3.63103971e-02 -8.00294816e-01 -1.40418696e+00 -6.76856101e-01 -3.19503248e-01 9.58703756e-01 3.09235454e-01 -4.67920899...
[5.8810715675354, -0.8258413076400757]
b2d3c932-8042-42cd-be12-bbcf33433479
frame-mining-a-free-lunch-for-learning
2210.07442
null
https://arxiv.org/abs/2210.07442v1
https://arxiv.org/pdf/2210.07442v1.pdf
Frame Mining: a Free Lunch for Learning Robotic Manipulation from 3D Point Clouds
We study how choices of input point cloud coordinate frames impact learning of manipulation skills from 3D point clouds. There exist a variety of coordinate frame choices to normalize captured robot-object-interaction point clouds. We find that different frames have a profound effect on agent learning performance, and ...
['Hao Su', 'Yangyan Li', 'Zhan Ling', 'Xuanlin Li', 'Minghua Liu']
2022-10-14
null
null
null
null
['robot-manipulation']
['robots']
[-2.14029938e-01 -3.81587356e-01 -4.59697664e-01 -2.69109488e-01 -4.91208673e-01 -7.42068172e-01 6.19089961e-01 -5.17936312e-02 -6.71355128e-01 3.37172598e-01 -1.67838082e-01 -1.60746932e-01 -1.85090140e-01 -3.99265379e-01 -1.11270761e+00 -5.59194386e-01 -1.69133693e-01 7.19582081e-01 6.09573722e-01 -4.15199906...
[4.821953773498535, 0.4014095664024353]
b0e30e51-07c6-4032-bbf8-98567fe43a48
adaensemble-learning-adaptively-sparse
2301.08353
null
https://arxiv.org/abs/2301.08353v1
https://arxiv.org/pdf/2301.08353v1.pdf
AdaEnsemble: Learning Adaptively Sparse Structured Ensemble Network for Click-Through Rate Prediction
Learning feature interactions is crucial to success for large-scale CTR prediction in recommender systems and Ads ranking. Researchers and practitioners extensively proposed various neural network architectures for searching and modeling feature interactions. However, we observe that different datasets favor different ...
['Liubo Li', 'YaChen Yan']
2023-01-06
null
null
null
null
['click-through-rate-prediction']
['miscellaneous']
[-3.63244891e-01 -5.64713895e-01 -5.61024010e-01 -7.95240223e-01 -3.80474061e-01 -4.54803884e-01 2.72376686e-01 -2.55672246e-01 -1.44325584e-01 2.87392259e-01 2.05530375e-01 -1.70464218e-01 -5.85528374e-01 -7.69126713e-01 -6.96033239e-01 -4.64330107e-01 -4.69150424e-01 8.43320847e-01 3.99514228e-01 -4.24452364...
[10.144776344299316, 5.484459400177002]
40743fa8-3301-48ea-a999-602d9f62d158
multi-timeline-summarization-mtls-improving
null
null
https://aclanthology.org/2021.acl-long.32
https://aclanthology.org/2021.acl-long.32.pdf
Multi-TimeLine Summarization (MTLS): Improving Timeline Summarization by Generating Multiple Summaries
In this paper, we address a novel task, Multiple TimeLine Summarization (MTLS), which extends the flexibility and versatility of Time-Line Summarization (TLS). Given any collection of time-stamped news articles, MTLS automatically discovers important yet different stories and generates a corresponding time-line for eac...
['Masatoshi Yoshikawa', 'Kazunari Sugiyama', 'Antoine Doucet', 'Adam Jatowt', 'Yi Yu']
2021-08-01
null
null
null
acl-2021-5
['timeline-summarization']
['natural-language-processing']
[ 2.96462744e-01 9.47682261e-02 -6.58264279e-01 -3.74529302e-01 -1.26728368e+00 -5.22481382e-01 6.84974313e-01 6.27805412e-01 -3.08137536e-01 1.04651034e+00 1.02574646e+00 1.38942540e-01 -1.29173055e-01 -5.32111287e-01 -6.58051968e-01 -3.15159947e-01 1.03930958e-01 6.22269154e-01 4.78463233e-01 -1.78463906...
[12.560020446777344, 9.485445022583008]
4c60af3d-ba6f-453f-ac34-06c5e6a48920
learning-sparse-auto-encoders-for-green-ai
2209.04448
null
https://arxiv.org/abs/2209.04448v1
https://arxiv.org/pdf/2209.04448v1.pdf
Learning sparse auto-encoders for green AI image coding
Recently, convolutional auto-encoders (CAE) were introduced for image coding. They achieved performance improvements over the state-of-the-art JPEG2000 method. However, these performances were obtained using massive CAEs featuring a large number of parameters and whose training required heavy computational power.\\ In ...
['Michel Barlaud', 'Marc Antonini', 'Frédéric Guyard', 'Cyprien Gille']
2022-09-09
null
null
null
null
['sparse-learning']
['methodology']
[ 1.79910645e-01 -1.13048486e-01 -2.29958519e-01 -1.18788801e-01 -5.16609788e-01 2.72385150e-01 -1.13434449e-01 -2.35590767e-02 -6.48955643e-01 7.47821033e-01 -3.27971689e-02 -1.53582990e-01 -2.77483881e-01 -8.05478036e-01 -7.34587848e-01 -7.88914502e-01 -2.08196193e-01 -8.27185586e-02 2.55456328e-01 -3.58334519...
[11.35383415222168, -1.7352826595306396]
98b7a31b-5f42-4ca9-8c85-481d5e6190c8
prior-information-based-decomposition-and
2303.01776
null
https://arxiv.org/abs/2303.01776v1
https://arxiv.org/pdf/2303.01776v1.pdf
Prior Information based Decomposition and Reconstruction Learning for Micro-Expression Recognition
Micro-expression recognition (MER) draws intensive research interest as micro-expressions (MEs) can infer genuine emotions. Prior information can guide the model to learn discriminative ME features effectively. However, most works focus on researching the general models with a stronger representation ability to adaptiv...
['Guoying Zhao', 'Yue Xie', 'Jingjie Yan', 'Guanming Lu', 'Haoyu Chen', 'Jinsheng Wei']
2023-03-03
null
null
null
null
['micro-expression-recognition']
['computer-vision']
[ 8.99796039e-02 1.17672369e-01 -4.24055785e-01 -5.82089782e-01 -9.48568359e-02 8.04501399e-02 5.29541016e-01 -4.01241720e-01 2.60988146e-01 -7.15042725e-02 5.57485044e-01 4.22341824e-01 -2.85444885e-01 -8.34412694e-01 -3.43347847e-01 -8.08765233e-01 -1.82868928e-01 -8.18598494e-02 -1.88988566e-01 -6.04570508...
[13.681722640991211, 1.5798745155334473]
5270d004-2aff-49fa-835d-d45d345dfcea
alternative-telescopic-displacement-an
2306.16950
null
https://arxiv.org/abs/2306.16950v1
https://arxiv.org/pdf/2306.16950v1.pdf
Alternative Telescopic Displacement: An Efficient Multimodal Alignment Method
Feature alignment is the primary means of fusing multimodal data. We propose a feature alignment method that fully fuses multimodal information, which alternately shifts and expands feature information from different modalities to have a consistent representation in a feature space. The proposed method can robustly cap...
['Xiaojun Zhang', 'Zong Lu', 'Zihong Luo Chengzhi Liu', 'Yitao Xu', 'Jiahao Qin']
2023-06-29
null
null
null
null
['arrhythmia-detection', 'cross-modal-retrieval', 'time-series-forecasting']
['medical', 'miscellaneous', 'time-series']
[ 1.07432753e-01 -5.63720345e-01 -2.80161828e-01 -5.13118207e-01 -1.44878566e+00 -5.70038080e-01 7.79902220e-01 2.33786449e-01 -2.95787573e-01 6.02808535e-01 6.11211598e-01 3.96247923e-01 -1.76983252e-01 -1.44096509e-01 -1.64277643e-01 -8.74099255e-01 9.31162667e-03 2.08466962e-01 -2.19106779e-01 5.78806736...
[13.138978004455566, 4.9800310134887695]
2a32ef0b-279e-4006-baa5-3cd3df96fffa
rgbd-salient-object-detection-via-deep-fusion
1607.03333
null
http://arxiv.org/abs/1607.03333v1
http://arxiv.org/pdf/1607.03333v1.pdf
RGBD Salient Object Detection via Deep Fusion
Numerous efforts have been made to design different low level saliency cues for the RGBD saliency detection, such as color or depth contrast features, background and color compactness priors. However, how these saliency cues interact with each other and how to incorporate these low level saliency cues effectively to ge...
['Shengfeng He', 'Jiandong Tian', 'Yandong Tang', 'Qingxiong Yang', 'Liangqiong Qu', 'Jiawei Zhang']
2016-07-12
null
null
null
null
['rgb-d-salient-object-detection']
['computer-vision']
[ 4.05681223e-01 8.79768282e-03 -2.37507209e-01 -3.32340777e-01 -4.54777628e-01 -1.10517412e-01 3.40518683e-01 1.90676004e-01 -3.22967887e-01 5.24749815e-01 2.77848154e-01 2.21116859e-02 7.27714673e-02 -6.10346735e-01 -5.94089925e-01 -6.01261199e-01 1.80894718e-01 -4.64905351e-01 9.99242544e-01 -2.45749295...
[9.784577369689941, -0.555191159248352]
6fdbbfdd-3590-4848-8dfb-b50139e3334d
adversarial-robustness-and-feature-impact
2303.13649
null
https://arxiv.org/abs/2303.13649v1
https://arxiv.org/pdf/2303.13649v1.pdf
Adversarial Robustness and Feature Impact Analysis for Driver Drowsiness Detection
Drowsy driving is a major cause of road accidents, but drivers are dismissive of the impact that fatigue can have on their reaction times. To detect drowsiness before any impairment occurs, a promising strategy is using Machine Learning (ML) to monitor Heart Rate Variability (HRV) signals. This work presents multiple e...
['André Lourenço', 'Isabel Praça', 'Eva Maia', 'Lourenço Rodrigues', 'João Vitorino']
2023-03-23
null
null
null
null
['heart-rate-variability']
['medical']
[-1.49292117e-02 1.40322044e-01 1.31092444e-01 -3.41498137e-01 -2.41466984e-01 -3.69926989e-01 1.94416881e-01 1.61977068e-01 -5.37806213e-01 8.20414543e-01 -7.36843497e-02 -4.56689537e-01 -2.80265480e-01 -5.68009734e-01 -4.10358012e-01 -7.77817190e-01 -2.23012567e-01 -2.76898444e-01 1.36359006e-01 -6.91948295...
[13.499878883361816, 2.8742592334747314]
e2f4c887-b001-4928-8691-08b09a3a4173
max-pooling-loss-training-of-long-short-term
1705.02411
null
http://arxiv.org/abs/1705.02411v1
http://arxiv.org/pdf/1705.02411v1.pdf
Max-Pooling Loss Training of Long Short-Term Memory Networks for Small-Footprint Keyword Spotting
We propose a max-pooling based loss function for training Long Short-Term Memory (LSTM) networks for small-footprint keyword spotting (KWS), with low CPU, memory, and latency requirements. The max-pooling loss training can be further guided by initializing with a cross-entropy loss trained network. A posterior smoothin...
['Geng-Shen Fu', 'Ming Sun', 'Anirudh Raju', 'Nikko Strom', 'George Tucker', 'Arindam Mandal', 'Spyros Matsoukas', 'Shiv Vitaladevuni', 'Sankaran Panchapagesan']
2017-05-05
null
null
null
null
['small-footprint-keyword-spotting']
['speech']
[ 1.45306855e-01 -2.51059681e-02 -4.11802560e-01 -4.88822281e-01 -1.01821589e+00 -2.77937222e-02 2.99729556e-01 -1.03119828e-01 -1.07073641e+00 7.15874314e-01 1.12856537e-01 -6.17425025e-01 1.44317016e-01 -5.34374177e-01 -8.68757069e-01 -5.47912657e-01 -1.59566984e-01 -8.15818235e-02 2.98053846e-02 1.25942126...
[14.19853687286377, 6.459183216094971]
1e1e3f6b-7248-47ea-842d-b4705723c371
multimodal-end-to-end-group-emotion
2111.05890
null
https://arxiv.org/abs/2111.05890v1
https://arxiv.org/pdf/2111.05890v1.pdf
Multimodal End-to-End Group Emotion Recognition using Cross-Modal Attention
Classifying group-level emotions is a challenging task due to complexity of video, in which not only visual, but also audio information should be taken into consideration. Existing works on multimodal emotion recognition are using bulky approach, where pretrained neural networks are used as a feature extractors and the...
['Lev Evtodienko']
2021-11-10
null
null
null
null
['multimodal-emotion-recognition', 'multimodal-emotion-recognition']
['computer-vision', 'speech']
[ 2.74958253e-01 -3.23077254e-02 1.56197757e-01 -4.76396084e-01 -6.30345523e-01 -5.36666572e-01 4.06381309e-01 2.60772742e-02 -7.93274522e-01 5.88223934e-01 2.80509919e-01 2.54293501e-01 1.33488998e-01 -4.34819549e-01 -6.52145267e-01 -6.13707483e-01 -1.51825622e-01 1.27958953e-02 6.49890769e-03 -1.29138932...
[13.266549110412598, 5.1063127517700195]
46a73860-86fb-4d10-84a4-23a00fe18a4c
fast-and-flexible-protein-design-using-deep
null
null
https://www.cell.com/cell-systems/fulltext/S2405-4712(20)30327-6
https://www.cell.com/action/showPdf?pii=S2405-4712%2820%2930327-6
Fast and Flexible Protein Design Using Deep Graph Neural Networks
Protein structure and function is determined by the arrangement of the linear sequence of amino acids in 3D space. We show that a deep graph neural network, ProteinSolver, can precisely design sequences that fold into a predetermined shape by phrasing this challenge as a constraint satisfaction problem (CSP), akin to S...
['Philip M. Kim', 'Albert Perez-Riba', 'Carles Corbi-Verge', 'David Becerra', 'Alexey Strokach']
2020-09-23
null
null
null
null
['protein-design']
['medical']
[-9.78067592e-02 1.66688189e-01 -1.51120439e-01 -3.23591471e-01 -2.66862094e-01 -1.03290355e+00 -1.41633630e-01 5.02319150e-02 -1.15583457e-01 1.31104493e+00 1.01902373e-01 -7.46751428e-01 6.59155175e-02 -4.97545958e-01 -1.33930731e+00 -6.71338320e-01 -1.95227191e-01 8.15604448e-01 -1.14416152e-01 -2.76340485...
[4.730828762054443, 5.606196403503418]
06569eb3-00b5-4998-8d64-22e003535bf4
relational-boosted-bandits
2012.09220
null
https://arxiv.org/abs/2012.09220v1
https://arxiv.org/pdf/2012.09220v1.pdf
Relational Boosted Bandits
Contextual bandits algorithms have become essential in real-world user interaction problems in recent years. However, these algorithms rely on context as attribute value representation, which makes them unfeasible for real-world domains like social networks are inherently relational. We propose Relational Boosted Bandi...
['Balaraman Ravindran', 'Sriraam Natarajan', 'Ashutosh Kakadiya']
2020-12-16
null
null
null
null
['explainable-models']
['computer-vision']
[ 5.06969579e-02 6.24503374e-01 -1.34399462e+00 -6.81800008e-01 -2.99691200e-01 -2.44728982e-01 4.93348360e-01 1.31284997e-01 2.55799741e-01 1.04931736e+00 4.90900487e-01 -8.03148270e-01 -7.82732308e-01 -1.11186516e+00 -8.71184707e-01 -2.25926638e-01 -2.37535685e-01 7.99399495e-01 4.87211496e-02 -3.54107171...
[8.888039588928223, 7.651078224182129]
e0ce67f9-ab58-4ba8-b2ae-668e5ba51a40
evolutionary-hierarchical-dirichlet-process
null
null
https://aclanthology.org/P13-2099
https://aclanthology.org/P13-2099.pdf
Evolutionary Hierarchical Dirichlet Process for Timeline Summarization
null
['Jiwei Li', 'Sujian Li']
2013-08-01
null
null
null
acl-2013-8
['timeline-summarization']
['natural-language-processing']
[-8.63703638e-02 1.71006292e-01 -6.22772932e-01 -4.08054382e-01 -8.41685571e-03 -9.08429027e-01 6.55310392e-01 -6.53472245e-01 -2.85945535e-01 1.06888819e+00 -4.63127941e-02 -1.01159286e+00 -3.91567826e-01 -9.63214397e-01 -4.95059669e-01 -6.31337762e-01 -9.79754329e-01 7.25764990e-01 3.30370307e-01 -6.93831444...
[-7.35828971862793, 3.7603485584259033]
7637d6cf-6c08-4f3a-81b9-9bf403be52fc
video-based-frame-level-facial-analysis-of
null
null
https://openaccess.thecvf.com/content/CVPR2022W/ABAW/html/Savchenko_Video-Based_Frame-Level_Facial_Analysis_of_Affective_Behavior_on_Mobile_Devices_CVPRW_2022_paper.html
https://openaccess.thecvf.com/content/CVPR2022W/ABAW/papers/Savchenko_Video-Based_Frame-Level_Facial_Analysis_of_Affective_Behavior_on_Mobile_Devices_CVPRW_2022_paper.pdf
Video-Based Frame-Level Facial Analysis of Affective Behavior on Mobile Devices Using EfficientNets
In this paper, we consider the problem of real-time video-based facial emotion analytics, namely, facial expression recognition, prediction of valence and arousal and detection of action unit points. We propose the novel frame-level emotion recognition algorithm by extracting facial features with the single EfficientNe...
['Savchenko A.V.']
2022-06-10
null
null
null
cvpr-workshop-2022-6
['facial-expression-recognition', 'action-unit-detection']
['computer-vision', 'computer-vision']
[ 2.28242591e-01 -1.87755432e-02 7.62001649e-02 -7.72823095e-01 -7.66865969e-01 -2.45775297e-01 4.51530188e-01 3.13562937e-02 -5.83633840e-01 5.19147277e-01 9.56002697e-02 6.07451379e-01 3.70827913e-01 -9.49034914e-02 -3.72689396e-01 -7.81389654e-01 -4.32289034e-01 -1.19334422e-01 -3.81213218e-01 -3.76383275...
[13.578600883483887, 2.0753636360168457]
012cadd0-f152-4b06-bae1-deb88e4f0352
three-dimensional-blind-image-deconvolution
1904.09974
null
http://arxiv.org/abs/1904.09974v1
http://arxiv.org/pdf/1904.09974v1.pdf
Three dimensional blind image deconvolution for fluorescence microscopy using generative adversarial networks
Due to image blurring image deconvolution is often used for studying biological structures in fluorescence microscopy. Fluorescence microscopy image volumes inherently suffer from intensity inhomogeneity, blur, and are corrupted by various types of noise which exacerbate image quality at deeper tissue depth. Therefore,...
['Shuo Han', 'Edward J. Delp', 'Paul Salama', 'Soonam Lee', 'Kenneth W. Dunn']
2019-04-19
null
null
null
null
['image-deconvolution']
['computer-vision']
[ 4.03108269e-01 -6.45393848e-01 7.73830116e-01 -1.78439140e-01 -4.70677733e-01 -8.27966869e-01 -9.16981697e-03 -4.34740067e-01 -8.00283670e-01 1.28226626e+00 1.05816759e-02 -4.88236845e-02 -5.38988505e-03 -2.27702763e-02 -3.87423933e-01 -1.39730072e+00 3.21723878e-01 1.98918320e-02 -3.73949222e-02 3.41600955...
[12.62283992767334, -2.665886402130127]
ac30f5b1-0a3c-48f8-bf21-ccc1a52f0f3d
semi-supervised-soil-moisture-prediction
2012.03506
null
https://arxiv.org/abs/2012.03506v2
https://arxiv.org/pdf/2012.03506v2.pdf
Dynamic Structure Learning through Graph Neural Network for Forecasting Soil Moisture in Precision Agriculture
Soil moisture is an important component of precision agriculture as it directly impacts the growth and quality of vegetation. Forecasting soil moisture is essential to schedule the irrigation and optimize the use of water. Physics based soil moisture models need rich features and heavy computation which is not scalable...
['Sambaran Bandyopadhyay', 'Anoushka Vyas']
2020-12-07
null
null
null
null
['graph-structure-learning']
['graphs']
[ 2.70013362e-01 -8.02743137e-02 -3.09668362e-01 -2.14339137e-01 3.35345715e-01 -4.32055175e-01 1.82385251e-01 7.86364734e-01 1.07213557e-02 8.50510299e-01 -3.15455645e-01 -9.76531863e-01 -3.89830112e-01 -1.74162531e+00 -5.89395165e-01 -6.94901943e-01 -5.34762621e-01 1.05181627e-01 5.65880656e-01 -5.96714675...
[9.380097389221191, -1.5409661531448364]
3b164b15-c1dc-4165-a12a-d432f5539f49
table-detection-for-visually-rich-document
2305.19181
null
https://arxiv.org/abs/2305.19181v1
https://arxiv.org/pdf/2305.19181v1.pdf
Table Detection for Visually Rich Document Images
Table Detection (TD) is a fundamental task towards visually rich document understanding. Current studies usually formulate the TD problem as an object detection problem, then leverage Intersection over Union (IoU) based metrics to evaluate the model performance and IoU-based loss functions to optimize the model. TD app...
['Ala Abu Alkheir', 'Burak Kantarci', 'Murat Simsek', 'Bin Xiao']
2023-05-30
null
null
null
null
['table-detection']
['miscellaneous']
[-7.84197822e-02 -1.71376541e-01 -2.85720229e-01 -2.53307223e-01 -8.04868102e-01 -3.60658646e-01 3.92658085e-01 3.09906095e-01 -1.09005466e-01 5.81296325e-01 3.60015512e-01 -1.17467448e-01 3.19552273e-02 -8.73337328e-01 -8.72143030e-01 -5.97993731e-01 2.56509721e-01 2.99460977e-01 4.56588835e-01 1.61920875...
[11.486440658569336, 2.305534601211548]
dea3bb65-eed4-4a73-9770-7a8002b82aa7
pythia-a-suite-for-analyzing-large-language
2304.01373
null
https://arxiv.org/abs/2304.01373v2
https://arxiv.org/pdf/2304.01373v2.pdf
Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
How do large language models (LLMs) develop and evolve over the course of training? How do these patterns change as models scale? To answer these questions, we introduce \textit{Pythia}, a suite of 16 LLMs all trained on public data seen in the exact same order and ranging in size from 70M to 12B parameters. We provide...
['Oskar van der Wal', 'Lintang Sutawika', 'Aviya Skowron', 'Edward Raff', 'USVSN Sai Prashanth', 'Shivanshu Purohit', 'Mohammad Aflah Khan', 'Eric Hallahan', "Kyle O'Brien", 'Herbie Bradley', 'Quentin Anthony', 'Hailey Schoelkopf', 'Stella Biderman']
2023-04-03
null
null
null
null
['memorization']
['natural-language-processing']
[-1.08689934e-01 -2.57384181e-01 -3.97861421e-01 -4.35543776e-01 -8.84861708e-01 -7.20521271e-01 6.93723261e-01 8.43022093e-02 -5.33954799e-01 6.10684812e-01 2.88065635e-02 -6.66076958e-01 6.87341094e-02 -2.96904624e-01 -8.57383907e-01 -4.67508763e-01 -2.09320247e-01 6.05766177e-01 1.31258428e-01 -1.67988092...
[10.646623611450195, 8.310639381408691]
da969a55-f3ac-4724-9153-4c01ecbef544
abcnet-v2-adaptive-bezier-curve-network-for
2105.03620
null
https://arxiv.org/abs/2105.03620v3
https://arxiv.org/pdf/2105.03620v3.pdf
ABCNet v2: Adaptive Bezier-Curve Network for Real-time End-to-end Text Spotting
End-to-end text-spotting, which aims to integrate detection and recognition in a unified framework, has attracted increasing attention due to its simplicity of the two complimentary tasks. It remains an open problem especially when processing arbitrarily-shaped text instances. Previous methods can be roughly categorize...
['Hao Chen', 'Chongyu Liu', 'Peng Chen', 'Tong He', 'Lianwen Jin', 'Chunhua Shen', 'Yuliang Liu']
2021-05-08
null
null
null
null
['text-spotting']
['computer-vision']
[ 4.21264261e-01 -5.42643249e-01 1.47010639e-01 -2.41218165e-01 -6.40538871e-01 -6.30686820e-01 6.64292157e-01 1.71513334e-01 -6.60748482e-01 2.03746930e-01 -1.12114184e-01 -4.92838621e-01 1.83079496e-01 -7.19374835e-01 -6.33661628e-01 -5.26775360e-01 5.16796172e-01 3.08617324e-01 3.23601902e-01 -2.31570065...
[12.035871505737305, 2.2355618476867676]
15ff7eec-4272-4631-88e7-e49f119816a1
teacher-student-network-for-3d-point-cloud
2210.17258
null
https://arxiv.org/abs/2210.17258v2
https://arxiv.org/pdf/2210.17258v2.pdf
Teacher-Student Network for 3D Point Cloud Anomaly Detection with Few Normal Samples
Anomaly detection, which is a critical and popular topic in computer vision, aims to detect anomalous samples that are different from the normal (i.e., non-anomalous) ones. The current mainstream methods focus on anomaly detection for images, whereas little attention has been paid to 3D point cloud. In this paper, draw...
['Chao Zhang', 'Jun Yu', 'Chunzhi Gu', 'Jianjian Qin']
2022-10-31
null
null
null
null
['3d-anomaly-detection']
['methodology']
[-3.60702202e-02 -1.21040586e-02 3.09140533e-01 -2.61108130e-01 -4.84459162e-01 -2.19577551e-01 5.54809809e-01 4.16150540e-01 -1.75779030e-01 -2.00205043e-01 -4.11527067e-01 -2.54885048e-01 6.46485314e-02 -5.13207734e-01 -6.55209303e-01 -8.31778288e-01 -1.19534083e-01 3.51747304e-01 4.64815766e-01 -1.30307779...
[7.633332252502441, 2.166276693344116]
2b7e088b-ce3a-4c4d-89ee-85253b238fab
evaluating-pre-trained-language-models-on
null
null
https://aclanthology.org/2022.sdp-1.22
https://aclanthology.org/2022.sdp-1.22.pdf
Evaluating Pre-Trained Language Models on Multi-Document Summarization for Literature Reviews
Systematic literature reviews in the biomedical space are often expensive to conduct. Automation through machine learning and large language models could improve the accuracy and research outcomes from such reviews. In this study, we evaluate a pre-trained LongT5 model on the MSLR22: Multi-Document Summarization for Li...
['Benjamin Yu']
null
null
null
null
sdp-coling-2022-10
['document-summarization']
['natural-language-processing']
[ 7.72917792e-02 1.77119106e-01 -6.92016423e-01 -2.60415226e-01 -1.69420266e+00 -4.72208142e-01 5.24410963e-01 8.15796971e-01 -3.86303216e-01 1.16907132e+00 1.07790899e+00 -6.37126982e-01 -2.85786480e-01 -1.00967944e-01 -6.08554006e-01 -4.64332709e-03 2.35038653e-01 4.41709578e-01 -3.12471837e-01 2.14033216...
[12.36037540435791, 9.604561805725098]
a8e9ee19-005f-4874-a407-a26b42caca13
measuring-faithful-and-plausible-visual
2305.15015
null
https://arxiv.org/abs/2305.15015v1
https://arxiv.org/pdf/2305.15015v1.pdf
Measuring Faithful and Plausible Visual Grounding in VQA
Metrics for Visual Grounding (VG) in Visual Question Answering (VQA) systems primarily aim to measure a system's reliance on relevant parts of the image when inferring an answer to the given question. Lack of VG has been a common problem among state-of-the-art VQA systems and can manifest in over-reliance on irrelevant...
['Tanja Schultz', 'Felix Putze', 'Daniel Reich']
2023-05-24
null
null
null
null
['visual-grounding']
['computer-vision']
[-4.69224714e-03 1.75214857e-01 1.37511846e-02 -4.50256407e-01 -7.26876318e-01 -8.27153921e-01 7.83099592e-01 3.68239433e-01 1.39909685e-01 4.08544332e-01 1.95692927e-01 -5.46754897e-01 -2.48802930e-01 -7.71354914e-01 -5.17599463e-01 -3.02044272e-01 5.78563869e-01 3.23431104e-01 3.90463918e-01 -4.79260713...
[10.92969036102295, 1.8096896409988403]
c9ac957a-6df8-4e36-b102-e77eb72b6230
self-eval-self-supervised-fine-grained
2208.08094
null
https://arxiv.org/abs/2208.08094v5
https://arxiv.org/pdf/2208.08094v5.pdf
SelF-Eval: Self-supervised Fine-grained Dialogue Evaluation
This paper introduces a novel Self-supervised Fine-grained Dialogue Evaluation framework (SelF-Eval). The core idea is to model the correlation between turn quality and the entire dialogue quality. We first propose a novel automatic data construction method that can automatically assign fine-grained scores for arbitrar...
['Ting Liu', 'Mingda Li', 'Weinan Zhang', 'Ziyu Zhuang', 'Longxuan Ma']
2022-08-17
null
https://aclanthology.org/2022.coling-1.39
https://aclanthology.org/2022.coling-1.39.pdf
coling-2022-10
['dialogue-evaluation']
['natural-language-processing']
[-3.67018521e-01 3.16921115e-01 -1.31778613e-01 -8.97847295e-01 -1.09996414e+00 -5.67643762e-01 8.02576125e-01 2.10952356e-01 -3.92597467e-01 1.09260511e+00 7.59357989e-01 4.44031432e-02 -1.56385019e-01 -6.73358858e-01 1.24529107e-02 -1.82310581e-01 1.88179791e-01 9.74890947e-01 2.99746990e-01 -9.20916677...
[12.767324447631836, 8.131373405456543]
2fb437b7-b6fd-4cd3-b9a1-20cd422c0c97
personalized-federated-recommender-systems
2212.08779
null
https://arxiv.org/abs/2212.08779v1
https://arxiv.org/pdf/2212.08779v1.pdf
Personalized Federated Recommender Systems with Private and Partially Federated AutoEncoders
Recommender Systems (RSs) have become increasingly important in many application domains, such as digital marketing. Conventional RSs often need to collect users' data, centralize them on the server-side, and form a global model to generate reliable recommendations. However, they suffer from two critical limitations: t...
['Jie Ding', 'Vahid Tarokh', 'Ali Anwar', 'Xinran Wang', 'Enmao Diao', 'Qi Le']
2022-12-17
null
null
null
null
['marketing']
['miscellaneous']
[-3.24880093e-01 5.50397597e-02 -3.93867016e-01 -7.64534116e-01 -7.85575986e-01 -4.14675176e-01 1.56213045e-01 -1.61458671e-01 -2.36448571e-01 4.78426248e-01 5.75956523e-01 -8.14811587e-02 -2.17831701e-01 -8.92632961e-01 -8.82882535e-01 -7.02062726e-01 -1.01706184e-01 4.09652531e-01 -2.33693078e-01 -2.16925547...
[5.9230546951293945, 6.293081760406494]
db797666-3100-46c6-b797-5547648da1b1
deep-representation-learning-for
null
null
https://www.sciencedirect.com/science/article/pii/S1532046419302229
https://www.sciencedirect.com/science/article/pii/S1532046419302229/pdfft?md5=b6a2b5ddc61984744f599ba097379d32&pid=1-s2.0-S1532046419302229-main.pdf
Deep representation learning for individualized treatment effect estimation using electronic health records
Utilizing clinical observational data to estimate individualized treatment effects (ITE) is a challenging task, as confounding inevitably exists in clinical data. Most of the existing models for ITE estimation tackle this problem by creating unbiased estimators of the treatment effects. Although valuable, learning a ba...
['Zhengxing Huang', 'Kunlun He', 'Uzay Kaymak', 'Xudong', 'Wei Dong', 'Peipei Chen']
2019-12-01
null
null
null
journal-of-biomedical-informatics-2019-12
['causal-inference', 'causal-inference']
['knowledge-base', 'miscellaneous']
[ 2.15343535e-01 3.63729708e-02 -8.77167046e-01 -5.44418395e-01 -1.09354186e+00 -1.10033907e-01 2.10497111e-01 2.32200980e-01 -2.66146868e-01 1.27520072e+00 6.80780530e-01 -2.78666973e-01 -6.85873508e-01 -5.78867972e-01 -6.25746191e-01 -7.90297866e-01 -3.41920525e-01 5.81243455e-01 -5.46365798e-01 4.88777131...
[7.998507976531982, 5.6124982833862305]
537818c9-3da0-4f33-89a7-aab940cff139
reflection-removal-using-low-rank-matrix
null
null
http://openaccess.thecvf.com/content_cvpr_2017/html/Han_Reflection_Removal_Using_CVPR_2017_paper.html
http://openaccess.thecvf.com/content_cvpr_2017/papers/Han_Reflection_Removal_Using_CVPR_2017_paper.pdf
Reflection Removal Using Low-Rank Matrix Completion
The images taken through glass often capture a target transmitted scene as well as undesired reflected scenes. In this paper, we propose a low-rank matrix completion algorithm to remove reflection artifacts automatically from multiple glass images taken at slightly different camera locations. We assume that the transmi...
['Jae-Young Sim', 'Byeong-Ju Han']
2017-07-01
null
null
null
cvpr-2017-7
['reflection-removal']
['computer-vision']
[ 0.66825014 -0.19038658 0.5519239 -0.20232275 -0.72354466 -0.10995226 -0.18761599 -0.7207479 -0.12585324 0.2822958 0.41321465 0.4610328 -0.2375103 -0.43803048 -0.6551302 -1.2409654 0.24336089 -0.17198001 0.20397963 -0.11654833 0.39437443 0.01245691 -1.2974501 0.48753113 0.7448243 0.78007966 0.6...
[10.370481491088867, -2.7857789993286133]
f2f4c8ca-2d76-41db-a63c-968b0a768a3f
real-or-fake-text-investigating-human-ability
2212.12672
null
https://arxiv.org/abs/2212.12672v1
https://arxiv.org/pdf/2212.12672v1.pdf
Real or Fake Text?: Investigating Human Ability to Detect Boundaries Between Human-Written and Machine-Generated Text
As text generated by large language models proliferates, it becomes vital to understand how humans engage with such text, and whether or not they are able to detect when the text they are reading did not originate with a human writer. Prior work on human detection of generated text focuses on the case where an entire p...
['Chris Callison-Burch', 'Sherry Shi', 'Arun Kirubarajan', 'Daphne Ippolito', 'Liam Dugan']
2022-12-24
null
null
null
null
['human-detection']
['computer-vision']
[ 2.96167403e-01 1.43018499e-01 9.85059440e-02 -2.58507848e-01 -1.02403283e+00 -8.23955655e-01 6.80593073e-01 7.33829260e-01 -8.25043678e-01 5.55854559e-01 4.84600037e-01 -4.30869192e-01 3.89816701e-01 -3.31335694e-01 -5.83634377e-01 -3.58205549e-02 5.14442861e-01 5.49555957e-01 7.46638253e-02 -1.93199903...
[11.708534240722656, 8.938375473022461]
2eeaf98b-13b9-442f-8451-678326c345e9
a-new-approach-for-pedestrian-density
1811.05006
null
https://arxiv.org/abs/1811.05006v2
https://arxiv.org/pdf/1811.05006v2.pdf
A new approach for pedestrian density estimation using moving sensors and computer vision
An understanding of pedestrian dynamics is indispensable for numerous urban applications including the design of transportation networks and planing for business development. Pedestrian counting often requires utilizing manual or technical means to count individuals in each location of interest. However, such methods d...
['Roberto M. Cesar-Jr.', 'Gabriel B. A. Ferreira', 'Eric K. Tokuda', 'David Boyle', 'Yitzchak Lockerman', 'Claudio T. Silva', 'Ethan Sorrelgreen']
2018-11-12
null
null
null
null
['pedestrian-density-estimation']
['computer-vision']
[-1.45080581e-01 -3.05344194e-01 7.01652616e-02 -1.72669411e-01 -5.08204877e-01 -3.15071017e-01 6.27047658e-01 1.87573627e-01 -6.86163127e-01 8.77452791e-01 -6.00249879e-02 -5.61008096e-01 1.05998717e-01 -1.17573357e+00 -4.78088737e-01 -5.59136689e-01 -8.75348300e-02 6.68120027e-01 4.86701429e-01 7.10818693...
[8.22743034362793, -0.43251267075538635]
3e651286-8776-428c-999d-f31da33f664b
focusing-on-what-to-decode-and-what-to-train
2307.02291
null
https://arxiv.org/abs/2307.02291v1
https://arxiv.org/pdf/2307.02291v1.pdf
Focusing on what to decode and what to train: Efficient Training with HOI Split Decoders and Specific Target Guided DeNoising
Recent one-stage transformer-based methods achieve notable gains in the Human-object Interaction Detection (HOI) task by leveraging the detection of DETR. However, the current methods redirect the detection target of the object decoder, and the box target is not explicitly separated from the query embeddings, which lea...
['Keiji Yanai', 'Yingcheng Wang', 'Junwen Chen']
2023-07-05
null
null
null
null
['human-object-interaction-detection', 'object-detection']
['computer-vision', 'computer-vision']
[ 3.52688372e-01 2.79192090e-01 -2.10769743e-01 -4.31470007e-01 -9.16515052e-01 -3.01797926e-01 3.33494753e-01 -1.20325349e-01 -4.77938831e-01 2.52716064e-01 1.58050403e-01 2.07358282e-02 2.31633395e-01 -4.81608719e-01 -9.24070537e-01 -5.37899375e-01 5.73438764e-01 5.67651808e-01 4.76882279e-01 1.52774706...
[9.624897003173828, 1.3614614009857178]
d6886628-fbcf-47a9-bb15-e9f9422024bd
dynamic-spatial-propagation-network-for-depth
2202.09769
null
https://arxiv.org/abs/2202.09769v1
https://arxiv.org/pdf/2202.09769v1.pdf
Dynamic Spatial Propagation Network for Depth Completion
Image-guided depth completion aims to generate dense depth maps with sparse depth measurements and corresponding RGB images. Currently, spatial propagation networks (SPNs) are the most popular affinity-based methods in depth completion, but they still suffer from the representation limitation of the fixed affinity and ...
['Hua Yang', 'Wending Zhou', 'Qi Zhong', 'Tao Cheng', 'Yuankai Lin']
2022-02-20
null
null
null
null
['depth-completion']
['computer-vision']
[ 7.77468681e-02 2.33120441e-01 9.43265706e-02 -2.85068363e-01 -5.12431204e-01 5.54339541e-03 6.01331413e-01 2.48803392e-01 -8.48802030e-01 6.22735202e-01 3.69970113e-01 1.48982629e-01 -2.60561913e-01 -1.06899917e+00 -7.16944337e-01 -7.49979615e-01 8.65292549e-02 7.72732794e-01 7.27737546e-01 -2.28443250...
[8.846720695495605, -2.3612003326416016]
f96eea17-a43c-4f90-a194-016f8af882bb
balancing-effect-of-training-dataset
2305.15582
null
https://arxiv.org/abs/2305.15582v1
https://arxiv.org/pdf/2305.15582v1.pdf
Balancing Effect of Training Dataset Distribution of Multiple Styles for Multi-Style Text Transfer
Text style transfer is an exciting task within the field of natural language generation that is often plagued by the need for high-quality paired datasets. Furthermore, training a model for multi-attribute text style transfer requires datasets with sufficient support across all combinations of the considered stylistic ...
['Dongyeop Kang', 'David Ma', 'Debarati Das']
2023-05-24
null
null
null
null
['style-transfer', 'text-style-transfoer']
['computer-vision', 'natural-language-processing']
[ 5.47194004e-01 -1.24694400e-01 -1.24411546e-01 -5.33540010e-01 -7.31706083e-01 -8.88724506e-01 7.49181509e-01 -1.02751860e-02 -4.61434782e-01 9.64516461e-01 4.70832497e-01 -2.64774173e-01 8.84776637e-02 -8.77701223e-01 -7.83268213e-01 -4.45750684e-01 5.02063274e-01 6.92168236e-01 -1.05622225e-01 -3.82600129...
[11.52079963684082, 9.725409507751465]
284e9333-4682-483b-9e64-33e05f26b640
csg0-continual-urban-scene-generation-with
2112.03252
null
https://arxiv.org/abs/2112.03252v2
https://arxiv.org/pdf/2112.03252v2.pdf
CSG0: Continual Urban Scene Generation with Zero Forgetting
With the rapid advances in generative adversarial networks (GANs), the visual quality of synthesised scenes keeps improving, including for complex urban scenes with applications to automated driving. We address in this work a continual scene generation setup in which GANs are trained on a stream of distinct domains; id...
['Matthieu Cord', 'Patrick Pérez', 'Tuan-Hung Vu', 'Himalaya Jain']
2021-12-06
null
null
null
null
['scene-generation']
['computer-vision']
[ 5.35594583e-01 3.50525022e-01 2.05029234e-01 8.80882051e-03 -6.14212275e-01 -5.87326765e-01 9.26789522e-01 -1.56893060e-01 -3.31341028e-01 1.18226540e+00 -3.43688689e-02 -1.53192237e-01 1.24482132e-01 -1.14122999e+00 -1.12261057e+00 -7.52138674e-01 3.35779160e-01 6.10008657e-01 2.64827996e-01 -2.63094425...
[11.679048538208008, -0.403053879737854]
33223977-a844-4a58-8942-44825b23f4c8
dfdl-discriminative-feature-oriented
1502.01032
null
http://arxiv.org/abs/1502.01032v1
http://arxiv.org/pdf/1502.01032v1.pdf
DFDL: Discriminative Feature-oriented Dictionary Learning for Histopathological Image Classification
In histopathological image analysis, feature extraction for classification is a challenging task due to the diversity of histology features suitable for each problem as well as presence of rich geometrical structure. In this paper, we propose an automatic feature discovery framework for extracting discriminative class-...
['Hojjat S. Mousavi', 'Vishal Monga', 'UK Arvind Rao', 'Ganesh Rao', 'Tiep H. Vu']
2015-02-03
null
null
null
null
['histopathological-image-classification']
['medical']
[ 1.75713316e-01 -3.20465773e-01 -1.80503696e-01 -2.01953158e-01 -9.18628633e-01 -4.06580895e-01 4.62368309e-01 7.61272490e-01 -4.03167844e-01 5.93104482e-01 9.27989110e-02 -2.31252149e-01 -5.08362651e-01 -6.19135499e-01 -9.75039080e-02 -1.34547007e+00 -4.56252009e-01 6.60040200e-01 2.91767806e-01 -2.57311221...
[15.057138442993164, -2.881216049194336]
9ed25fbe-ac36-46d8-a2fa-54012b63c0f2
load-local-orientation-adaptive-descriptor
1504.05809
null
http://arxiv.org/abs/1504.05809v1
http://arxiv.org/pdf/1504.05809v1.pdf
LOAD: Local Orientation Adaptive Descriptor for Texture and Material Classification
In this paper, we propose a novel local feature, called Local Orientation Adaptive Descriptor (LOAD), to capture regional texture in an image. In LOAD, we proposed to define point description on an Adaptive Coordinate System (ACS), adopt a binary sequence descriptor to capture relationships between one point and its ne...
['Xianbiao Qi', 'Qingquan Li', 'Linlin Shen', 'Matti Pietikainen', 'Guoying Zhao']
2015-04-22
null
null
null
null
['material-classification', 'material-recognition']
['computer-vision', 'computer-vision']
[ 2.59546936e-01 -6.66133344e-01 -2.07578093e-01 -2.98594624e-01 -6.15737319e-01 -3.13404381e-01 6.61381662e-01 5.27741946e-02 -4.87058252e-01 4.32239711e-01 -9.89510417e-02 2.62462407e-01 -4.21977222e-01 -9.04350102e-01 -5.47198176e-01 -1.06938100e+00 -6.36531487e-02 -1.01524033e-01 4.02896672e-01 -3.69139344...
[10.382233619689941, -0.3696097433567047]
a8cb55a1-5473-4992-8acd-cbe5e8a7e025
faster-tad-towards-temporal-action-detection
2204.02674
null
https://arxiv.org/abs/2204.02674v1
https://arxiv.org/pdf/2204.02674v1.pdf
Faster-TAD: Towards Temporal Action Detection with Proposal Generation and Classification in a Unified Network
Temporal action detection (TAD) aims to detect the semantic labels and boundaries of action instances in untrimmed videos. Current mainstream approaches are multi-step solutions, which fall short in efficiency and flexibility. In this paper, we propose a unified network for TAD, termed Faster-TAD, by re-purposing a Fas...
['Yandong Guo', 'Xunqiang Tao', 'Wei Li', 'Chen Chen', 'Shimin Chen']
2022-04-06
null
null
null
null
['action-spotting']
['computer-vision']
[ 2.63949394e-01 -9.81742069e-02 -4.92651194e-01 -1.31268755e-01 -5.64289808e-01 -3.65916997e-01 5.51163197e-01 -4.46267188e-01 -5.39411187e-01 4.89811063e-01 5.25860965e-01 5.93279526e-02 1.31336600e-01 -5.13836503e-01 -4.63609248e-01 -5.79016745e-01 -1.02015778e-01 -2.45520864e-02 1.05875027e+00 -2.21543938...
[8.383606910705566, 0.5133983492851257]
ddb63b45-c1af-4540-98a3-32b958ba668d
sources-of-uncertainty-in-machine-learning-a
2305.16703
null
https://arxiv.org/abs/2305.16703v1
https://arxiv.org/pdf/2305.16703v1.pdf
Sources of Uncertainty in Machine Learning -- A Statisticians' View
Machine Learning and Deep Learning have achieved an impressive standard today, enabling us to answer questions that were inconceivable a few years ago. Besides these successes, it becomes clear, that beyond pure prediction, which is the primary strength of most supervised machine learning algorithms, the quantification...
['Göran Kauermann', 'Frauke Kreuter', 'Malte Schierholz', 'Patrick Oliver Schenk', 'Cornelia Gruber']
2023-05-26
null
null
null
null
['miscellaneous']
['miscellaneous']
[-2.03512073e-01 4.18959051e-01 -1.45653710e-01 -6.09567702e-01 -7.67640889e-01 -5.35734236e-01 1.02618873e+00 5.03686368e-01 -5.08146286e-01 9.87626672e-01 2.30995014e-01 -4.74385560e-01 -7.86131322e-01 -9.37094808e-01 -6.07983589e-01 -7.87678540e-01 -4.95787710e-02 4.96260673e-01 -1.66535154e-01 1.61689714...
[7.588131904602051, 4.09501838684082]
862ceeba-d9fd-4474-90e3-2bac1c4c8bd2
learning-summary-worthy-visual-representation
2305.04824
null
https://arxiv.org/abs/2305.04824v1
https://arxiv.org/pdf/2305.04824v1.pdf
Learning Summary-Worthy Visual Representation for Abstractive Summarization in Video
Multimodal abstractive summarization for videos (MAS) requires generating a concise textual summary to describe the highlights of a video according to multimodal resources, in our case, the video content and its transcript. Inspired by the success of the large-scale generative pre-trained language model (GPLM) in gener...
['Qun Liu', 'Xin Jiang', 'Zexuan Qiu', 'Qinliang Su', 'Yasheng Wang', 'Xiaojun Meng', 'Zenan Xu']
2023-05-08
null
null
null
null
['abstractive-text-summarization']
['natural-language-processing']
[ 4.06497717e-01 -2.70736264e-03 -2.72978306e-01 -2.11447701e-01 -1.39641154e+00 -6.05136096e-01 7.07105994e-01 1.09627936e-02 -1.64086193e-01 7.40912259e-01 8.80246580e-01 9.15913209e-02 4.85878557e-01 -3.48869890e-01 -9.50329840e-01 -6.71594381e-01 2.35548675e-01 -8.85755867e-02 -6.24409690e-02 -5.19902147...
[10.666755676269531, 0.6948280930519104]
420ba012-c8f8-4dc1-a2ac-ccbc73d1e686
a-deep-dive-into-dataset-imbalance-and-bias
2203.08235
null
https://arxiv.org/abs/2203.08235v1
https://arxiv.org/pdf/2203.08235v1.pdf
A Deep Dive into Dataset Imbalance and Bias in Face Identification
As the deployment of automated face recognition (FR) systems proliferates, bias in these systems is not just an academic question, but a matter of public concern. Media portrayals often center imbalance as the main source of bias, i.e., that FR models perform worse on images of non-white people or women because these d...
['Tom Goldstein', 'Micah Goldblum', 'Hossein Souri', 'Samuel Dooley', 'Steven Reich', 'Valeriia Cherepanova']
2022-03-15
null
null
null
null
['face-identification']
['computer-vision']
[ 0.29770455 -0.13551442 -0.30469325 -0.5444574 -0.30794957 -0.57206786 0.31736684 0.09171268 -0.36740765 0.53002465 0.33392057 -0.44135243 -0.02198026 -0.62931406 -0.44256976 -0.63981014 0.30663994 0.27402744 -0.53328943 -0.03414468 0.40655977 0.589834 -1.6167222 0.07096032 0.5521152 0.7903514 -0....
[12.984983444213867, 1.2172410488128662]
a95f2cca-a31a-47f8-b057-48a8a009b0b9
enhancing-visual-domain-adaptation-with
2306.10142
null
https://arxiv.org/abs/2306.10142v1
https://arxiv.org/pdf/2306.10142v1.pdf
Enhancing Visual Domain Adaptation with Source Preparation
Robotic Perception in diverse domains such as low-light scenarios, where new modalities like thermal imaging and specialized night-vision sensors are increasingly employed, remains a challenge. Largely, this is due to the limited availability of labeled data. Existing Domain Adaptation (DA) techniques, while promising ...
['Jeff Schneider', 'Christoph Mertz', 'Anurag Ghosh', 'Anirudha Ramesh']
2023-06-16
null
null
null
null
['unsupervised-domain-adaptation']
['methodology']
[ 7.02680647e-01 -2.62471437e-01 -2.00573444e-01 -5.71600854e-01 -9.94585037e-01 -9.96069312e-01 5.85524559e-01 -3.71461987e-01 -6.16745830e-01 5.89038372e-01 -2.98021510e-02 1.31443009e-01 2.87313983e-02 -3.31007987e-01 -7.75189877e-01 -1.00036287e+00 5.19479275e-01 5.31956792e-01 5.44028044e-01 -1.83212042...
[8.701216697692871, -1.993804931640625]
8646372d-b2cc-4bd1-a3d9-f1e34f314921
superpixel-perception-graph-neural-network
2210.07539
null
https://arxiv.org/abs/2210.07539v1
https://arxiv.org/pdf/2210.07539v1.pdf
Superpixel Perception Graph Neural Network for Intelligent Defect Detection
Aero-engine is the core component of aircraft and other spacecraft. The high-speed rotating blades provide power by sucking in air and fully combusting, and various defects will inevitably occur, threatening the operation safety of aero-engine. Therefore, regular inspections are essential for such a complex system. How...
['Ruqiang Yan', 'Xuefeng Chen', 'Chuang Sun', 'Qixiu Yang', 'Hongbing Shang']
2022-10-14
null
null
null
null
['defect-detection']
['computer-vision']
[ 2.85483059e-02 -7.73321614e-02 2.89936215e-01 1.38711348e-01 1.21028975e-01 -2.93816060e-01 6.00661784e-02 -2.47855857e-01 1.30373031e-01 2.36100271e-01 -3.32036525e-01 -3.60957384e-01 -6.31094053e-02 -1.14985943e+00 -4.62490678e-01 -8.67211044e-01 -1.62761565e-02 -9.13203508e-02 4.43145663e-01 -3.11961979...
[9.603386878967285, -0.8162592053413391]
ebd9836d-fd97-4ed7-a42c-d173b31742fa
understanding-user-behavior-in-carousel
2307.01866
null
https://arxiv.org/abs/2307.01866v1
https://arxiv.org/pdf/2307.01866v1.pdf
Understanding User Behavior in Carousel Recommendation Systems for Click Modeling and Learning to Rank
Carousels (also-known as multilists) have become the standard user interface for e-commerce platforms replacing the ranked list, the previous standard for recommender systems. While the research community has begun to focus on carousels, there are many unanswered questions and undeveloped areas when compared to the lit...
['Santiago de Leon-Martinez']
2023-07-04
null
null
null
null
['learning-to-rank', 'learning-to-rank', 'movie-recommendation', 'information-retrieval']
['graphs', 'miscellaneous', 'miscellaneous', 'natural-language-processing']
[-4.77145314e-02 -1.18615001e-01 -6.10293508e-01 -4.48285282e-01 -7.41634220e-02 -8.58483374e-01 4.46220040e-01 1.33454455e-02 -4.54999596e-01 1.08442418e-01 2.50684589e-01 -9.11332846e-01 -7.79919386e-01 -4.39598411e-01 -2.57525206e-01 -3.15233655e-02 -4.95421030e-02 2.02991113e-01 4.80032533e-01 -4.24286276...
[10.044998168945312, 5.806351661682129]
d7adf715-2d1f-4aaf-a761-d805e71cf728
keyword-aware-influential-community-search-in
1912.02114
null
https://arxiv.org/abs/1912.02114v1
https://arxiv.org/pdf/1912.02114v1.pdf
Keyword Aware Influential Community Search in Large Attributed Graphs
We introduce a novel keyword-aware influential community query KICQ that finds the most influential communities from an attributed graph, where an influential community is defined as a closely connected group of vertices having some dominance over other groups of vertices with the expertise (a set of keywords) matching...
['Farhana M. Choudhury', 'Yong-Bin Kang', 'Md. Saiful Islam', 'Mohammed Eunus Ali', 'Timos Sellis']
2019-12-04
null
null
null
null
['community-search']
['graphs']
[-7.43508488e-02 9.61382464e-02 -2.38834172e-01 5.06709963e-02 -2.23205194e-01 -7.21978188e-01 7.64720559e-01 7.69468188e-01 -4.37715679e-01 2.87992835e-01 4.33965355e-01 -6.56128749e-02 -7.23981917e-01 -1.25406039e+00 -2.90433258e-01 -4.76370186e-01 -5.46753287e-01 6.95768833e-01 7.30264783e-01 -2.55844355...
[7.304025173187256, 5.878608226776123]
c3ee841f-94a5-40b7-b2d4-af4effc4b538
towards-dense-volumetric-pancreas
1711.06439
null
http://arxiv.org/abs/1711.06439v2
http://arxiv.org/pdf/1711.06439v2.pdf
Towards dense volumetric pancreas segmentation in CT using 3D fully convolutional networks
Pancreas segmentation in computed tomography imaging has been historically difficult for automated methods because of the large shape and size variations between patients. In this work, we describe a custom-build 3D fully convolutional network (FCN) that can process a 3D image including the whole pancreas and produce a...
['Kensaku MORI', 'Kazunari Misawa', 'Hirohisa ODA', 'Yuichiro Hayashi', 'Takayuki Kitasaka', 'Michitaka Fujiwara', 'Holger Roth', 'Masahiro Oda', 'Natsuki Shimizu']
2017-11-17
null
null
null
null
['pancreas-segmentation']
['medical']
[ 5.96094877e-02 2.15445504e-01 6.85702786e-02 -4.29654151e-01 -5.66547751e-01 -6.69684470e-01 1.89050540e-01 2.53853053e-01 -4.95060146e-01 3.58247042e-01 2.28324607e-01 -6.84286118e-01 1.34864952e-02 -3.92881066e-01 -6.95517957e-01 -6.23936594e-01 -6.90145433e-01 9.04730439e-01 9.61918384e-02 2.97175109...
[14.494793891906738, -2.610658645629883]
961b169e-01e2-42df-aed4-c3ba71739846
inversion-of-1d-frequency-and-time-domain
1912.00612
null
https://arxiv.org/abs/1912.00612v1
https://arxiv.org/pdf/1912.00612v1.pdf
Inversion of 1D frequency- and time-domain electromagnetic data with convolutional neural networks
Inversion of electromagnetic data finds applications in many areas of geophysics. The inverse problem is commonly solved with either deterministic optimization methods (such as the nonlinear conjugate gradient or Gauss-Newton) which are prone to getting trapped in a local minimum, or probabilistic methods which are ver...
['Andrei Swidinsky', 'Vladimir Puzyrev']
2019-12-02
null
null
null
null
['geophysics']
['miscellaneous']
[ 8.72159600e-02 -2.11218253e-01 4.87744242e-01 -4.98720378e-01 -7.97832549e-01 -4.84897941e-01 6.01703107e-01 1.05257519e-01 -6.85223579e-01 1.12369812e+00 -3.06298733e-01 -7.77985632e-01 -5.47234893e-01 -1.07248640e+00 -7.71982193e-01 -9.62566912e-01 -2.99166620e-01 7.24525273e-01 -5.83264083e-02 -5.78354657...
[6.709049701690674, 2.8370156288146973]
686205ba-3d0b-4166-a65a-7673213a0f1e
pc2-projection-conditioned-point-cloud
null
null
http://openaccess.thecvf.com//content/CVPR2023/html/Melas-Kyriazi_PC2_Projection-Conditioned_Point_Cloud_Diffusion_for_Single-Image_3D_Reconstruction_CVPR_2023_paper.html
http://openaccess.thecvf.com//content/CVPR2023/papers/Melas-Kyriazi_PC2_Projection-Conditioned_Point_Cloud_Diffusion_for_Single-Image_3D_Reconstruction_CVPR_2023_paper.pdf
PC2: Projection-Conditioned Point Cloud Diffusion for Single-Image 3D Reconstruction
Reconstructing the 3D shape of an object from a single RGB image is a long-standing problem in computer vision. In this paper, we propose a novel method for single-image 3D reconstruction which generates a sparse point cloud via a conditional denoising diffusion process. Our method takes as input a single RGB image...
['Andrea Vedaldi', 'Christian Rupprecht', 'Luke Melas-Kyriazi']
2023-01-01
null
null
null
cvpr-2023-1
['3d-reconstruction']
['computer-vision']
[ 5.05649745e-01 -1.30674437e-01 4.52406794e-01 -1.48859158e-01 -9.86514926e-01 -7.29742467e-01 7.32040763e-01 -1.04122661e-01 -1.97279438e-01 2.59316921e-01 -6.12502545e-02 1.06141761e-01 5.21981232e-02 -1.09553075e+00 -1.13875961e+00 -8.98568809e-01 6.69801474e-01 1.14991236e+00 1.70447335e-01 1.82001591...
[8.7964448928833, -3.3324708938598633]
f1c15332-1217-4b16-822b-cb987fa114cb
ji-yu-duo-tou-zhu-yi-li-he-bilstmgai-jin
null
null
https://aclanthology.org/2020.ccl-1.30
https://aclanthology.org/2020.ccl-1.30.pdf
基于多头注意力和BiLSTM改进DAM模型的中文问答匹配方法(Chinese question answering method based on multi-head attention and BiLSTM improved DAM model)
针对目前检索式多轮对话深度注意力机制模型DAM(Deep Attention Matching Network)候选回复细节不匹配和语义混淆的问题,本文提出基于多头注意力和双向长短时记忆网络(BiLSTM)改进DAM模型的中文问答匹配方法,该方法采用多头注意力机制,使模型有能力建模较长的多轮对话,更好的处理目标回复与上下文的匹配关系。此外,本文在特征融合过程中采用BiLSTM模型,通过捕获多轮对话中的序列依赖关系,进一步提升选择目标候选回复的准确率。本文在豆瓣和电商两个开放数据集上进行实验,实验性能均优于DAM基线模型,R10@1指标在含有词向量增强的情况下提升了1.5%。
['Xia Zhao', 'Weijie Jiang', 'Chongchong Yu', 'Hanzhong Qin']
null
null
null
null
ccl-2020-10
['deep-attention', 'deep-attention']
['computer-vision', 'natural-language-processing']
[-0.81465846 -0.53620666 0.57658917 0.63949645 -0.03455365 -0.08120593 -0.06266747 1.298765 -0.3264723 0.166293 0.8044529 0.11080449 0.04242913 -1.2629861 -0.3989925 -1.2384853 -0.7696726 1.6158489 1.2553394 -0.91242003 0.29202372 0.9180892 -0.8747693 0.36349005 0.8449636 0.7818563 0.8...
[-3.3160908222198486, 6.907607555389404]
3e694808-1caf-4722-ba68-c9012edbad2c
improving-interpretability-via-regularization
2211.08686
null
https://arxiv.org/abs/2211.08686v1
https://arxiv.org/pdf/2211.08686v1.pdf
Improving Interpretability via Regularization of Neural Activation Sensitivity
State-of-the-art deep neural networks (DNNs) are highly effective at tackling many real-world tasks. However, their wide adoption in mission-critical contexts is hampered by two major weaknesses - their susceptibility to adversarial attacks and their opaqueness. The former raises concerns about the security and general...
['Asaf Shabtai', 'Ron Bitton', 'Gil Fidel', 'Ofir Moshe']
2022-11-16
null
null
null
null
['explanation-fidelity-evaluation', 'interpretability-techniques-for-deep-learning']
['methodology', 'miscellaneous']
[ 7.05161318e-02 3.87040645e-01 4.53269869e-01 -3.75833392e-01 -2.22631305e-01 -8.91219914e-01 7.11884260e-01 -2.08330646e-01 -6.34029329e-01 6.48503363e-01 8.30142796e-02 -6.80915833e-01 -7.90506676e-02 -6.20366156e-01 -9.53537166e-01 -5.20610213e-01 -1.09770276e-01 1.26297280e-01 1.49183556e-01 -6.03806734...
[5.678408622741699, 7.841391086578369]
b5066723-fbbd-4cb2-b05b-06bdb3a819a2
sub-pixel-face-landmarks-using-heatmaps-and-a
2103.03059
null
https://arxiv.org/abs/2103.03059v2
https://arxiv.org/pdf/2103.03059v2.pdf
Sub-pixel face landmarks using heatmaps and a bag of tricks
Accurate face landmark localization is an essential part of face recognition, reconstruction and morphing. To accurately localize face landmarks, we present our heatmap regression approach. Each model consists of a MobileNetV2 backbone followed by several upscaling layers, with different tricks to optimize both perform...
['Siwa Boonpunmongkol', 'Pavit Noinongyao', 'Sanjana Jain', 'Aubin Samacoits', 'Samuel W. F. Earp']
2021-03-04
null
null
null
null
['face-alignment']
['computer-vision']
[-2.67340094e-01 2.04422791e-02 -4.11386713e-02 -7.15173304e-01 -1.04410684e+00 -6.25511289e-01 6.42669678e-01 -2.72775412e-01 -5.56761444e-01 2.92355478e-01 2.76216511e-02 2.84405444e-02 3.16709548e-01 -3.45944136e-01 -1.07616591e+00 -5.05403399e-01 -3.74643505e-01 7.02858090e-01 4.63965870e-02 8.89592469...
[13.493762969970703, 0.33908575773239136]
9627b965-49d6-4345-8cd4-b1be94166e55
deep-semantics-aware-photo-adjustment
1706.08260
null
http://arxiv.org/abs/1706.08260v1
http://arxiv.org/pdf/1706.08260v1.pdf
Deep Semantics-Aware Photo Adjustment
Automatic photo adjustment is to mimic the photo retouching style of professional photographers and automatically adjust photos to the learned style. There have been many attempts to model the tone and the color adjustment globally with low-level color statistics. Also, spatially varying photo adjustment methods have b...
['Seonghyeon Nam', 'Seon Joo Kim']
2017-06-26
null
null
null
null
['photo-retouching']
['computer-vision']
[ 2.83648700e-01 -2.00563222e-01 -4.97634038e-02 -8.35198045e-01 -4.06123430e-01 -4.28220361e-01 4.40784216e-01 1.08446933e-01 -4.62966919e-01 1.69191927e-01 2.80062079e-01 2.70548254e-01 -3.90410013e-02 -7.92158604e-01 -7.10263252e-01 -5.61223090e-01 6.37862384e-01 -9.82572064e-02 3.47484201e-01 -4.44742978...
[11.366817474365234, -0.9952687621116638]
305968ef-a7c3-45b7-b194-91a2267641db
latentkeypointgan-controlling-gans-via-latent
2103.15812
null
https://arxiv.org/abs/2103.15812v4
https://arxiv.org/pdf/2103.15812v4.pdf
LatentKeypointGAN: Controlling GANs via Latent Keypoints
Generative adversarial networks (GANs) have attained photo-realistic quality in image generation. However, how to best control the image content remains an open challenge. We introduce LatentKeypointGAN, a two-stage GAN which is trained end-to-end on the classical GAN objective with internal conditioning on a set of sp...
['Helge Rhodin', 'Bastian Wandt', 'Xingzhe He']
2021-03-29
latentkeypointgan-controlling-gans-via-latent-1
https://openreview.net/forum?id=y_tIL5vki1l
https://openreview.net/pdf?id=y_tIL5vki1l
null
['unsupervised-facial-landmark-detection']
['computer-vision']
[ 4.69359875e-01 3.54855478e-01 1.31624117e-02 -7.98916742e-02 -7.36323118e-01 -9.45706844e-01 9.56222594e-01 -4.93284851e-01 -2.21640263e-02 5.44860363e-01 1.12909645e-01 1.09797217e-01 4.85846773e-02 -8.55522513e-01 -9.54476953e-01 -9.84938085e-01 4.16543245e-01 3.97458404e-01 -3.26637119e-01 -1.15001708...
[11.724371910095215, -0.48492559790611267]
88a1c00b-5a1a-4845-a398-b325e40c2fa2
deepfake-network-architecture-attribution
2202.13843
null
https://arxiv.org/abs/2202.13843v2
https://arxiv.org/pdf/2202.13843v2.pdf
Deepfake Network Architecture Attribution
With the rapid progress of generation technology, it has become necessary to attribute the origin of fake images. Existing works on fake image attribution perform multi-class classification on several Generative Adversarial Network (GAN) models and obtain high accuracies. While encouraging, these works are restricted t...
['Xirong Li', 'Lei LI', 'Juan Cao', 'Ziyao Huang', 'Tianyun Yang']
2022-02-28
null
null
null
null
['fake-image-attribution']
['computer-vision']
[ 4.78482664e-01 1.61816761e-01 -6.02790639e-02 -4.17049855e-01 -6.77661836e-01 -8.35907638e-01 6.98623776e-01 -5.23068130e-01 6.93793371e-02 7.97997653e-01 -2.36666456e-01 -3.65214646e-01 3.64603698e-01 -7.25652635e-01 -1.06126308e+00 -6.60302639e-01 3.12764943e-01 4.83424157e-01 5.02768643e-02 -2.09674329...
[12.520745277404785, 1.0452675819396973]
2684c6b7-4968-4bda-8d4d-062e5cfb802c
omdet-language-aware-object-detection-with
2209.05946
null
https://arxiv.org/abs/2209.05946v1
https://arxiv.org/pdf/2209.05946v1.pdf
OmDet: Language-Aware Object Detection with Large-scale Vision-Language Multi-dataset Pre-training
Advancing object detection to open-vocabulary and few-shot transfer has long been a challenge for computer vision research. This work explores a continual learning approach that enables a detector to expand its zero/few-shot capabilities via multi-dataset vision-language pre-training. Using natural language as knowledg...
['Kyusong Lee', 'Xiaopeng Lu', 'Peng Liu', 'Tiancheng Zhao']
2022-09-10
null
null
null
null
['open-vocabulary-object-detection']
['computer-vision']
[ 8.61147940e-02 -1.05756432e-01 -1.83674812e-01 -1.68295071e-01 -1.05648375e+00 -4.15734798e-01 7.57321417e-01 -1.72387674e-01 -9.49168503e-01 4.19665962e-01 -7.58544952e-02 -9.78412107e-02 2.95380265e-01 -4.15807545e-01 -7.08311558e-01 -4.61140484e-01 3.77350986e-01 4.92286593e-01 8.20143461e-01 -2.93580055...
[9.596418380737305, 1.492775321006775]
ba42f51a-2de1-44b0-a208-29d29e976a4d
a-survey-on-causal-discovery-methods-for
2303.15027
null
https://arxiv.org/abs/2303.15027v2
https://arxiv.org/pdf/2303.15027v2.pdf
A Survey on Causal Discovery Methods for Temporal and Non-Temporal Data
Causal Discovery (CD) is the process of identifying the cause-effect relationships among the variables of a system from data. Over the years, several methods have been developed primarily based on the statistical properties of data to uncover the underlying causal mechanism. In this study, we present an extensive discu...
['Md Osman Gani', 'Emam Hossain', 'Uzma Hasan']
2023-03-27
null
null
null
null
['causal-discovery']
['knowledge-base']
[ 3.26714069e-01 -1.09718762e-01 -4.11913067e-01 -3.34495604e-01 -3.67819279e-01 -5.58303475e-01 8.69299531e-01 3.07168990e-01 4.71869648e-01 1.05820572e+00 4.85175729e-01 -6.34731710e-01 -1.06320369e+00 -1.02500105e+00 -5.24354517e-01 -6.28625512e-01 -1.13565922e+00 3.84588748e-01 1.20154366e-01 2.45328844...
[7.835007667541504, 5.359640121459961]
b7eb60d1-73fb-4995-9260-eeeb989762c1
adaptive-knowledge-sharing-in-multi-task
null
null
https://aclanthology.org/P18-2104
https://aclanthology.org/P18-2104.pdf
Adaptive Knowledge Sharing in Multi-Task Learning: Improving Low-Resource Neural Machine Translation
Neural Machine Translation (NMT) is notorious for its need for large amounts of bilingual data. An effective approach to compensate for this requirement is Multi-Task Learning (MTL) to leverage different linguistic resources as a source of inductive bias. Current MTL architectures are based on the Seq2Seq transduction,...
['Poorya Zaremoodi', 'Gholamreza Haffari', 'Wray Buntine']
2018-07-01
null
null
null
acl-2018-7
['low-resource-neural-machine-translation']
['natural-language-processing']
[ 3.43385667e-01 -1.49698019e-01 -3.67037654e-01 -4.28758085e-01 -1.42345977e+00 -8.91872466e-01 7.90892601e-01 -2.85107642e-01 -6.46617115e-01 1.03246808e+00 6.03570700e-01 -8.39666069e-01 3.11699450e-01 -2.20610470e-01 -9.61669624e-01 -4.47614551e-01 3.33509833e-01 8.35646808e-01 -1.07792318e-01 -4.06641096...
[11.604574203491211, 10.188304901123047]
35ed3768-7c3b-42b6-8bc9-f92b430650ca
learning-to-measure-the-point-cloud
null
null
http://openaccess.thecvf.com//content/CVPR2023/html/Huang_Learning_To_Measure_the_Point_Cloud_Reconstruction_Loss_in_a_CVPR_2023_paper.html
http://openaccess.thecvf.com//content/CVPR2023/papers/Huang_Learning_To_Measure_the_Point_Cloud_Reconstruction_Loss_in_a_CVPR_2023_paper.pdf
Learning To Measure the Point Cloud Reconstruction Loss in a Representation Space
For point cloud reconstruction-related tasks, the reconstruction losses to evaluate the shape differences between reconstructed results and the ground truths are typically used to train the task networks. Most existing works measure the training loss with point-to-point distance, which may introduce extra defects a...
['Yong liu', 'Chengjie Wang', 'Mingang Chen', 'Zhenyu Zhang', 'Ying Tai', 'Jiangning Zhang', 'Zhonggan Ding', 'Tianxin Huang']
2023-01-01
null
null
null
cvpr-2023-1
['point-cloud-reconstruction']
['computer-vision']
[ 3.71617936e-02 -6.02239966e-02 1.93011731e-01 -4.21879351e-01 -7.36421347e-01 -3.54094863e-01 3.84532958e-01 -8.66338387e-02 -2.13088036e-01 4.01345313e-01 -2.48491302e-01 8.12393799e-02 -2.27338523e-01 -1.21403074e+00 -1.07414186e+00 -6.18026137e-01 1.69313908e-01 5.57156622e-01 3.52274060e-01 -3.09400201...
[8.379057884216309, -3.4644813537597656]
5d6f450a-3f0b-4148-937e-964d4a1311d7
multi-target-regression-via-output-space
2003.09896
null
https://arxiv.org/abs/2003.09896v1
https://arxiv.org/pdf/2003.09896v1.pdf
Multi-target regression via output space quantization
Multi-target regression is concerned with the prediction of multiple continuous target variables using a shared set of predictors. Two key challenges in multi-target regression are: (a) modelling target dependencies and (b) scalability to large output spaces. In this paper, a new multi-target regression method is propo...
['Ioannis Vlahavas', 'Konstantinos Sechidis', 'Eleftherios Spyromitros-Xioufis']
2020-03-22
null
null
null
null
['multi-target-regression']
['miscellaneous']
[ 4.98896807e-01 -1.71526670e-01 -4.50012565e-01 -4.92259324e-01 -1.39773023e+00 -2.99529463e-01 6.46045446e-01 2.84586579e-01 -1.21160783e-01 7.19763696e-01 -1.76423401e-01 -7.68340379e-02 -4.00938839e-01 -7.94403613e-01 -5.21916509e-01 -9.08499658e-01 1.70374848e-02 7.35286534e-01 6.87370673e-02 -2.10111976...
[9.08290958404541, 4.229432582855225]
6ca444ab-a1bf-428f-ba84-ee9a346eae3c
improving-word-mover-s-distance-by-leveraging
2211.06229
null
https://arxiv.org/abs/2211.06229v1
https://arxiv.org/pdf/2211.06229v1.pdf
Improving word mover's distance by leveraging self-attention matrix
Measuring the semantic similarity between two sentences is still an important task. The word mover's distance (WMD) computes the similarity via the optimal alignment between the sets of word embeddings. However, WMD does not utilize word order, making it difficult to distinguish sentences with large overlaps of similar...
['Hidetoshi Shimodaira', 'Sho Yokoi', 'Hiroaki Yamagiwa']
2022-11-11
null
null
null
null
['paraphrase-identification']
['natural-language-processing']
[-8.05305615e-02 -3.07793587e-01 -1.51411101e-01 -3.56937975e-01 -6.53401673e-01 -4.51834142e-01 3.66032481e-01 6.16546988e-01 -6.55991614e-01 1.76974013e-01 6.57143176e-01 -3.08580548e-01 -3.69741052e-01 -7.15599597e-01 -3.64417881e-02 -5.64868748e-01 3.52633506e-01 7.67403319e-02 2.24772573e-01 -2.83892632...
[10.770050048828125, 8.722293853759766]
39c0bf9e-2eb7-4d16-9694-4b1fd0809c12
kgpool-dynamic-knowledge-graph-context
2106.00459
null
https://arxiv.org/abs/2106.00459v2
https://arxiv.org/pdf/2106.00459v2.pdf
KGPool: Dynamic Knowledge Graph Context Selection for Relation Extraction
We present a novel method for relation extraction (RE) from a single sentence, mapping the sentence and two given entities to a canonical fact in a knowledge graph (KG). Especially in this presumed sentential RE setting, the context of a single sentence is often sparse. This paper introduces the KGPool method to addres...
['Vijay Saraswat', 'Saeedeh Shekarpour', 'Johannes Hoffart', "Isaiah Onando Mulang'", 'Kuldeep Singh', 'Anson Bastos', 'Abhishek Nadgeri']
2021-06-01
null
https://aclanthology.org/2021.findings-acl.48
https://aclanthology.org/2021.findings-acl.48.pdf
findings-acl-2021-8
['relationship-extraction-distant-supervised']
['natural-language-processing']
[ 1.77121371e-01 8.67431700e-01 -4.77859735e-01 -4.22272444e-01 -4.25775349e-01 -4.05313075e-01 6.13765240e-01 9.18271482e-01 -6.44090772e-01 1.20596492e+00 4.66424108e-01 -3.19070071e-01 -1.35726333e-01 -1.24942684e+00 -1.07203126e+00 -1.81591120e-02 -3.61560404e-01 5.84217250e-01 3.17703158e-01 -1.66325539...
[9.316832542419434, 8.414727210998535]
fb96dc9b-c3be-4718-9fd5-56621f37a784
towards-end-to-end-prosody-transfer-for
1803.09047
null
http://arxiv.org/abs/1803.09047v1
http://arxiv.org/pdf/1803.09047v1.pdf
Towards End-to-End Prosody Transfer for Expressive Speech Synthesis with Tacotron
We present an extension to the Tacotron speech synthesis architecture that learns a latent embedding space of prosody, derived from a reference acoustic representation containing the desired prosody. We show that conditioning Tacotron on this learned embedding space results in synthesized audio that matches the prosody...
['Ron J. Weiss', 'RJ Skerry-Ryan', 'Ying Xiao', 'Daisy Stanton', 'Yuxuan Wang', 'Rif A. Saurous', 'Eric Battenberg', 'Rob Clark', 'Joel Shor']
2018-03-24
towards-end-to-end-prosody-transfer-for-1
https://icml.cc/Conferences/2018/Schedule?showEvent=2281
http://proceedings.mlr.press/v80/skerry-ryan18a/skerry-ryan18a.pdf
icml-2018-7
['expressive-speech-synthesis']
['speech']
[ 1.24578506e-01 2.25087881e-01 -2.46865168e-01 -1.83730751e-01 -1.18494463e+00 -7.36543298e-01 4.74581689e-01 -4.89310831e-01 1.53507501e-01 5.23573875e-01 1.13131523e+00 4.22189683e-02 1.82736024e-01 -4.63260323e-01 -6.03574455e-01 -7.36233711e-01 1.08888231e-01 2.95537889e-01 -1.83885068e-01 -3.16061020...
[15.011167526245117, 6.555191516876221]
0d1550b4-89c4-40b1-9e91-212e2fd0242d
specializing-multilingual-language-models-an
2106.09063
null
https://arxiv.org/abs/2106.09063v4
https://arxiv.org/pdf/2106.09063v4.pdf
Specializing Multilingual Language Models: An Empirical Study
Pretrained multilingual language models have become a common tool in transferring NLP capabilities to low-resource languages, often with adaptations. In this work, we study the performance, extensibility, and interaction of two such adaptations: vocabulary augmentation and script transliteration. Our evaluations on par...
['Noah A. Smith', 'Ethan C. Chau']
2021-06-16
null
https://aclanthology.org/2021.mrl-1.5
https://aclanthology.org/2021.mrl-1.5.pdf
emnlp-mrl-2021-11
['transliteration', 'pretrained-multilingual-language-models']
['natural-language-processing', 'natural-language-processing']
[-2.25396112e-01 -9.64756534e-02 -6.45547032e-01 -4.87667352e-01 -9.40041363e-01 -1.15318215e+00 5.07628500e-01 7.53718102e-03 -8.25778544e-01 1.04883623e+00 4.76344913e-01 -8.21033418e-01 4.33461219e-01 -3.47104907e-01 -6.21004283e-01 -3.67693906e-03 6.79419041e-02 6.65103734e-01 1.51104003e-01 -2.61838019...
[10.458941459655762, 9.943202018737793]
f29fcb12-cc22-48c7-b0ec-d8ff0ebf8b5c
mmea-entity-alignment-for-multi-modal
null
null
https://link.springer.com/chapter/10.1007/978-3-030-55130-8_12
http://home.ustc.edu.cn/~liyichen/assets/files/LiyiChen_KSEM20.pdf
MMEA: Entity Alignment for Multi-Modal Knowledge Graphs
Entity alignment plays an essential role in the knowledge graph (KG) integration. Though large efforts have been made on exploring the association of relational embeddings between different knowledge graphs, they may fail to effectively describe and integrate the multimodal knowledge in the real application scenario. T...
['Enhong Chen', 'Zhefeng Wang', 'Tong Xu', 'Yijun Wang', 'Zhi Li', 'Liyi Chen']
2020-08-20
null
null
null
null
['multi-modal-entity-alignment']
['knowledge-base']
[-3.38131756e-01 4.15427610e-02 -3.41483831e-01 -1.21537752e-01 -6.88671291e-01 -5.38126767e-01 5.56838095e-01 4.35973227e-01 -6.57595620e-02 3.03635597e-01 4.67706710e-01 -7.34659582e-02 -3.78332973e-01 -1.01382923e+00 -4.96138543e-01 -5.71990252e-01 2.17366174e-01 1.37071371e-01 -2.77176183e-02 -3.20030123...
[8.674450874328613, 7.728212833404541]
a0dc17cf-01d0-4974-9b53-8ccb1e691861
on-projection-methods-for-functional-time
2105.04399
null
https://arxiv.org/abs/2105.04399v1
https://arxiv.org/pdf/2105.04399v1.pdf
On projection methods for functional time series forecasting
Two nonparametric methods are presented for forecasting functional time series (FTS). The FTS we observe is a curve at a discrete-time point. We address both one-step-ahead forecasting and dynamic updating. Dynamic updating is a forward prediction of the unobserved segment of the most recent curve. Among the two propos...
['Hanlin Shang', 'Raúl Jiménez', 'Antonio Elías']
2021-05-10
null
null
null
null
['univariate-time-series-forecasting']
['time-series']
[ 1.49121985e-01 -1.62883028e-01 -1.71298981e-02 -3.86288673e-01 -3.94614607e-01 -5.49908638e-01 6.28492057e-01 1.00590251e-01 -4.71415780e-02 9.94338691e-01 1.30670354e-01 -3.68946642e-01 -5.41636407e-01 -9.41785812e-01 -7.51999617e-01 -1.10700715e+00 -3.69923115e-01 3.58655274e-01 9.60531458e-02 -3.05689186...
[6.852479934692383, 3.2490885257720947]
9234517b-bf4a-46cd-8513-edfefc2feb55
the-many-faces-of-1-lipschitz-neural-networks
2104.05097
null
https://arxiv.org/abs/2104.05097v6
https://arxiv.org/pdf/2104.05097v6.pdf
Pay attention to your loss: understanding misconceptions about 1-Lipschitz neural networks
Lipschitz constrained networks have gathered considerable attention in the deep learning community, with usages ranging from Wasserstein distance estimation to the training of certifiably robust classifiers. However they remain commonly considered as less accurate, and their properties in learning are still not fully u...
['Alberto González-Sanz', 'Corentin Friedrich', 'Mathieu Serrurier', 'Thibaut Boissin', 'Franck Mamalet', 'Louis Béthune']
2021-04-11
null
null
null
null
['misconceptions']
['miscellaneous']
[ 7.70585537e-02 1.38199791e-01 -4.50092226e-01 -6.76023483e-01 -7.36938477e-01 -7.13703334e-01 3.59394848e-01 1.33889422e-01 -7.35213935e-01 9.52081323e-01 -2.16880292e-01 -1.18497908e-01 -5.65019965e-01 -6.19974732e-01 -9.29455698e-01 -1.03267324e+00 -2.20896795e-01 2.49677479e-01 7.28262216e-02 -6.46862984...
[7.878880500793457, 3.8396413326263428]
c4ac8ee3-eeee-4ed0-bca1-0e70dc9038fe
cliff-carrying-location-information-in-full
2208.00571
null
https://arxiv.org/abs/2208.00571v2
https://arxiv.org/pdf/2208.00571v2.pdf
CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation
Top-down methods dominate the field of 3D human pose and shape estimation, because they are decoupled from human detection and allow researchers to focus on the core problem. However, cropping, their first step, discards the location information from the very beginning, which makes themselves unable to accurately predi...
['Youliang Yan', 'Songcen Xu', 'Zhensong Zhang', 'Jianzhuang Liu', 'Zhihao LI']
2022-08-01
null
null
null
null
['3d-human-pose-and-shape-estimation', 'unsupervised-3d-human-pose-estimation']
['computer-vision', 'computer-vision']
[-2.46507198e-01 -3.52490656e-02 -1.55847654e-01 -2.29330122e-01 -7.32218802e-01 -5.28808177e-01 3.27825725e-01 -3.22726816e-01 -3.80776316e-01 4.01504934e-01 4.70289826e-01 2.23631620e-01 3.70131612e-01 -4.10573334e-01 -7.21731603e-01 -5.73678374e-01 3.49144220e-01 7.37637877e-01 3.25886421e-02 -2.96806782...
[7.06966495513916, -0.889286458492279]
80ca3600-f4a9-46e2-9b8a-1a3672691259
classi-fly-inferring-aircraft-categories-from
1908.01061
null
https://arxiv.org/abs/1908.01061v2
https://arxiv.org/pdf/1908.01061v2.pdf
Classi-Fly: Inferring Aircraft Categories from Open Data using Machine Learning
In recent years, air traffic communication data has become easy to access, enabling novel research in many fields. Exploiting this new data source, a wide range of applications have emerged, from weather forecasting to stock market prediction, or the collection of information about military and government movements. Ty...
['Ivan Martinovic', 'Vincent Lenders', 'Matthew Smith', 'Martin Strohmeier']
2019-07-30
null
null
null
null
['stock-market-prediction']
['time-series']
[-1.77699938e-01 -2.97163457e-01 -5.66946208e-01 -2.77008504e-01 -5.49366653e-01 -1.00029194e+00 6.51998162e-01 5.96668720e-01 -4.56976593e-01 7.53373921e-01 3.32476497e-01 -5.20841062e-01 -3.46343786e-01 -1.10490906e+00 -5.18282473e-01 -2.41809472e-01 -3.60544980e-01 5.98265886e-01 3.19172412e-01 -6.50777593...
[7.409072399139404, 3.1328537464141846]
50a76c64-8091-4872-838c-3fc0019b2c51
music-composition-with-deep-learning-a-review
2108.12290
null
https://arxiv.org/abs/2108.12290v2
https://arxiv.org/pdf/2108.12290v2.pdf
Music Composition with Deep Learning: A Review
Generating a complex work of art such as a musical composition requires exhibiting true creativity that depends on a variety of factors that are related to the hierarchy of musical language. Music generation have been faced with Algorithmic methods and recently, with Deep Learning models that are being used in other fi...
['Jose R. Beltran', 'Carlos Hernandez-Olivan']
2021-08-27
null
null
null
null
['music-generation', 'music-generation']
['audio', 'music']
[ 1.90684408e-01 1.50727943e-01 3.13873082e-01 2.91556239e-01 1.65504396e-01 -8.02003205e-01 9.30964351e-01 -3.39329481e-01 -1.34742439e-01 3.63013595e-01 4.26887184e-01 6.99909702e-02 -6.12067044e-01 -9.68731105e-01 -3.47628295e-01 -5.88671327e-01 8.50318223e-02 8.40636075e-01 -2.92203903e-01 -7.72048473...
[16.08775520324707, 5.516983985900879]
67fb8b98-4a67-472e-8a9b-b1f0722c3001
lightweight-multi-drone-detection-and-3d
2202.09097
null
https://arxiv.org/abs/2202.09097v1
https://arxiv.org/pdf/2202.09097v1.pdf
Lightweight Multi-Drone Detection and 3D-Localization via YOLO
In this work, we present and evaluate a method to perform real-time multiple drone detection and three-dimensional localization using state-of-the-art tiny-YOLOv4 object detection algorithm and stereo triangulation. Our computer vision approach eliminates the need for computationally expensive stereo matching algorithm...
['Mangal Kothari', 'Nitik Jain', 'Aryan Sharma']
2022-02-18
null
null
null
null
['stereo-matching-1']
['computer-vision']
[-8.83405954e-02 -4.00702417e-01 7.09321856e-01 -1.72675192e-01 -3.88048559e-01 -1.03364503e+00 3.59054476e-01 3.51012908e-02 -7.44156599e-01 2.83981532e-01 -7.17079937e-01 -2.42301390e-01 2.71651536e-01 -8.49168837e-01 -5.99917948e-01 -2.68991023e-01 -5.94718099e-01 5.61352015e-01 9.36506987e-01 -3.55737865...
[8.27503776550293, -1.2591378688812256]
bf041729-5535-4e98-8380-68df420023d7
improving-iterative-text-revision-by-learning
2212.01350
null
https://arxiv.org/abs/2212.01350v1
https://arxiv.org/pdf/2212.01350v1.pdf
Improving Iterative Text Revision by Learning Where to Edit from Other Revision Tasks
Iterative text revision improves text quality by fixing grammatical errors, rephrasing for better readability or contextual appropriateness, or reorganizing sentence structures throughout a document. Most recent research has focused on understanding and classifying different types of edits in the iterative revision pro...
['Dongyeop Kang', 'Dhruv Kumar', 'Vipul Raheja', 'Wanyu Du', 'Zae Myung Kim']
2022-12-02
null
null
null
null
['grammatical-error-correction']
['natural-language-processing']
[ 8.12640429e-01 5.93751848e-01 -1.40434466e-02 -6.36870027e-01 -7.78570592e-01 -4.64362562e-01 5.55235505e-01 8.13886583e-01 -4.14051682e-01 6.96613848e-01 7.73164034e-01 -4.67761129e-01 -2.00041570e-02 -3.81917655e-01 -6.10408843e-01 5.32561481e-01 7.12021351e-01 5.90230346e-01 -1.82853535e-01 -6.62890911...
[11.902214050292969, 9.33910846710205]
2b67b8f8-04b3-4bcb-b33b-1a18f89baa79
deep-scattering-spectrum-germaneness-to-fault
2210.09837
null
https://arxiv.org/abs/2210.09837v3
https://arxiv.org/pdf/2210.09837v3.pdf
Deep Scattering Spectrum germaneness to Fault Detection and Diagnosis for Component-level Prognostics and Health Management (PHM)
In fault detection and diagnosis of prognostics and health management (PHM) systems, most of the methodologies utilize machine learning (ML) or deep learning (DL) through which either some features are extracted beforehand (in the case of ML) or filters are used to extract features autonomously (in case of DL) to perfo...
['Ali Rohan']
2022-10-18
null
null
null
null
['fault-detection', 'industrial-robots']
['miscellaneous', 'robots']
[ 8.18889886e-02 -1.81817397e-01 4.97970849e-01 -1.18518462e-02 -3.14656556e-01 2.55320203e-02 3.39652359e-01 2.65855163e-01 1.07314341e-01 6.37192011e-01 -4.26609904e-01 -1.60783917e-01 -7.88607478e-01 -6.90176189e-01 -4.07053143e-01 -9.89019513e-01 -4.19256270e-01 3.76222014e-01 1.57313675e-01 -2.59147733...
[6.753728866577148, 2.3765451908111572]
18d97aa1-9f40-4cc4-878b-89827b3c1de4
temporally-guided-articulated-hand-pose
2101.04281
null
https://arxiv.org/abs/2101.04281v2
https://arxiv.org/pdf/2101.04281v2.pdf
Temporally Guided Articulated Hand Pose Tracking in Surgical Videos
Articulated hand pose tracking is an under-explored problem that carries the potential for use in an extensive number of applications, especially in the medical domain. With a robust and accurate tracking system on in-vivo surgical videos, the motion dynamics and movement patterns of the hands can be captured and analy...
['Jason J. Corso', 'Donald S. Likosky', 'Francis D. Pagani', 'Milisa Manojlovich', 'Roger D. Dias', 'Steven J. Yule', 'Luowei Zhou', 'Nathan Louis']
2021-01-12
null
null
null
null
['skills-assessment']
['computer-vision']
[-1.15217730e-01 -1.71367124e-01 -6.32996023e-01 1.96436018e-01 -1.12630785e+00 -8.48918855e-01 1.94248885e-01 -1.23833247e-01 -5.72080076e-01 5.35635948e-01 4.99896914e-01 -9.47160646e-02 -3.03093970e-01 1.64085656e-01 -5.30143440e-01 -8.18119824e-01 -3.95449698e-01 5.75775802e-01 2.58053571e-01 4.03193757...
[14.02945613861084, -3.3355019092559814]
d27a280a-bb92-4891-8601-cae1f9ee8b46
multi-view-deep-subspace-clustering-networks
1908.01978
null
https://arxiv.org/abs/1908.01978v1
https://arxiv.org/pdf/1908.01978v1.pdf
Multi-view Deep Subspace Clustering Networks
Multi-view subspace clustering aims to discover the inherent structure by fusing multi-view complementary information. Most existing methods first extract multiple types of hand-crafted features and then learn a joint affinity matrix for clustering. The disadvantage lies in two aspects: 1) Multi-view relations are not ...
['QinGhua Hu', 'Dawei Du', 'Longyin Wen', 'Binyuan Hui', 'Pengfei Zhu', 'Changqing Zhang']
2019-08-06
null
null
null
null
['multi-view-subspace-clustering']
['computer-vision']
[-3.26243967e-01 -3.62722129e-01 -2.14323863e-01 -5.01459301e-01 -5.50728977e-01 -7.29648948e-01 5.10803819e-01 -5.61631143e-01 -2.68639643e-02 1.11840159e-01 4.39725608e-01 3.10073078e-01 -3.26031834e-01 -4.66695905e-01 -4.91248429e-01 -9.66041088e-01 2.29408890e-01 4.75848794e-01 -2.61250973e-01 -1.63872056...
[8.374654769897461, 4.542566776275635]
6853f3d9-891f-47c0-ad18-2e4faa5c2d50
exploring-the-impact-of-noise-and
2211.07445
null
https://arxiv.org/abs/2211.07445v1
https://arxiv.org/pdf/2211.07445v1.pdf
Exploring the Impact of Noise and Degradations on Heart Sound Classification Models
The development of data-driven heart sound classification models has been an active area of research in recent years. To develop such data-driven models in the first place, heart sound signals need to be captured using a signal acquisition device. However, it is almost impossible to capture noise-free heart sound signa...
['Susan Mckeever', 'Andrew Hines', 'Davoud Shariat Panah']
2022-11-14
null
null
null
null
['sound-classification']
['audio']
[ 2.89031267e-01 -2.27223188e-01 4.35303718e-01 -9.55911279e-02 -4.90850121e-01 -4.71796453e-01 8.03212076e-02 1.97474137e-01 8.08195770e-02 3.48521978e-01 4.88205254e-01 -3.40736389e-01 -5.22941887e-01 -6.59847260e-01 -2.03619182e-01 -8.18654895e-01 4.67391172e-03 -1.06058039e-01 1.81542695e-01 -1.70912564...
[14.290766716003418, 3.312272071838379]
fda42197-ce7f-41ec-82ef-7798e213924f
reference-based-video-super-resolution-using
2203.14537
null
https://arxiv.org/abs/2203.14537v1
https://arxiv.org/pdf/2203.14537v1.pdf
Reference-based Video Super-Resolution Using Multi-Camera Video Triplets
We propose the first reference-based video super-resolution (RefVSR) approach that utilizes reference videos for high-fidelity results. We focus on RefVSR in a triple-camera setting, where we aim at super-resolving a low-resolution ultra-wide video utilizing wide-angle and telephoto videos. We introduce the first RefVS...
['Seungyong Lee', 'Sunghyun Cho', 'Myeonghee Lee', 'Junyong Lee']
2022-03-28
null
http://openaccess.thecvf.com//content/CVPR2022/html/Lee_Reference-Based_Video_Super-Resolution_Using_Multi-Camera_Video_Triplets_CVPR_2022_paper.html
http://openaccess.thecvf.com//content/CVPR2022/papers/Lee_Reference-Based_Video_Super-Resolution_Using_Multi-Camera_Video_Triplets_CVPR_2022_paper.pdf
cvpr-2022-1
['video-super-resolution', 'reference-based-video-super-resolution']
['computer-vision', 'computer-vision']
[ 3.69200110e-01 -6.16742730e-01 -1.53124154e-01 -2.06186324e-01 -1.33524692e+00 -2.55323142e-01 5.77122271e-01 -8.61065388e-01 -2.65621215e-01 8.52156401e-01 4.98294502e-01 2.36621082e-01 -3.56786132e-01 -4.05113548e-01 -9.24669862e-01 -4.67574418e-01 -1.31704271e-01 -2.51884997e-01 5.50588071e-01 -2.96542972...
[11.004864692687988, -1.9430314302444458]
dcb8eac9-fc12-4b5c-b89a-dc97d5dcea39
a-new-information-theory-of-certainty-for
2304.12833
null
https://arxiv.org/abs/2304.12833v1
https://arxiv.org/pdf/2304.12833v1.pdf
A New Information Theory of Certainty for Machine Learning
Claude Shannon coined entropy to quantify the uncertainty of a random distribution for communication coding theory. We observe that the uncertainty nature of entropy also limits its direct usage in mathematical modeling. Therefore we propose a new concept troenpy,as the canonical dual of entropy, to quantify the certai...
['Arthur Jun Zhang']
2023-04-25
null
null
null
null
['document-classification']
['natural-language-processing']
[ 2.25513846e-01 1.74454868e-01 -1.33082017e-01 -1.60380512e-01 -5.11577010e-01 -6.29410625e-01 9.03914034e-01 2.70058513e-01 -5.03676355e-01 8.77439559e-01 1.48442954e-01 -5.46189129e-01 -2.48214141e-01 -9.24871027e-01 -2.66508609e-01 -9.60357785e-01 -2.72466332e-01 2.70855278e-01 -3.11408788e-01 -2.66794205...
[5.635475158691406, 4.939809799194336]
6ea3dac7-e3dd-4ddf-9cad-4b8207763a65
audio-to-intent-using-acoustic-textual
2210.12134
null
https://arxiv.org/abs/2210.12134v1
https://arxiv.org/pdf/2210.12134v1.pdf
Audio-to-Intent Using Acoustic-Textual Subword Representations from End-to-End ASR
Accurate prediction of the user intent to interact with a voice assistant (VA) on a device (e.g. on the phone) is critical for achieving naturalistic, engaging, and privacy-centric interactions with the VA. To this end, we present a novel approach to predict the user's intent (the user speaking to the device or not) di...
['Ahmed Tewfik', 'Xiaochuan Niu', 'Erik Marchi', 'Oggi Rudovic', 'Prateeth Nayak', 'Pranay Dighe']
2022-10-21
null
null
null
null
['intent-classification']
['natural-language-processing']
[ 4.91247118e-01 1.69856861e-01 -1.33353949e-01 -2.86601245e-01 -1.40921342e+00 -6.68698251e-01 3.36222947e-01 5.14352368e-03 -1.14644237e-01 1.73127815e-01 8.22729886e-01 -4.61841613e-01 1.28220990e-01 -2.43341163e-01 -5.49387395e-01 -4.45087016e-01 4.66019809e-02 1.48998141e-01 -2.66768754e-01 -8.43399689...
[14.631667137145996, 6.2125115394592285]
fb029d96-2b8f-4968-9abe-b2414bc28a18
edge-cloud-collaborative-learning-with
2304.05871
null
https://arxiv.org/abs/2304.05871v1
https://arxiv.org/pdf/2304.05871v1.pdf
Edge-cloud Collaborative Learning with Federated and Centralized Features
Federated learning (FL) is a popular way of edge computing that doesn't compromise users' privacy. Current FL paradigms assume that data only resides on the edge, while cloud servers only perform model averaging. However, in real-life situations such as recommender systems, the cloud server has the ability to store his...
['Chao Wu', 'Guannan Zhang', 'Wenliang Zhong', 'Yi Zhou', 'Qunwei Li', 'Zexi Li']
2023-04-12
null
null
null
null
['edge-computing']
['time-series']
[-6.04812086e-01 -1.72813669e-01 -5.32856405e-01 -2.61580408e-01 -3.31452250e-01 -6.29191518e-01 2.16942519e-01 -9.82500762e-02 -1.51687622e-01 6.20615423e-01 -3.99648063e-02 -5.38599432e-01 -2.85003573e-01 -8.58645976e-01 -5.31951249e-01 -5.07302463e-01 -2.10068733e-01 5.67159466e-02 -9.65883806e-02 2.03801230...
[5.897154808044434, 6.231959819793701]
13550ee7-84fb-4fd5-abac-26a067ae98a0
audio-visual-efficient-conformer-for-robust
2301.01456
null
https://arxiv.org/abs/2301.01456v1
https://arxiv.org/pdf/2301.01456v1.pdf
Audio-Visual Efficient Conformer for Robust Speech Recognition
End-to-end Automatic Speech Recognition (ASR) systems based on neural networks have seen large improvements in recent years. The availability of large scale hand-labeled datasets and sufficient computing resources made it possible to train powerful deep neural networks, reaching very low Word Error Rate (WER) on academ...
['Radu Timofte', 'Maxime Burchi']
2023-01-04
null
null
null
null
['robust-speech-recognition']
['speech']
[ 3.36843789e-01 1.37701035e-01 -1.21488914e-01 -3.36847007e-01 -1.45400739e+00 -1.37636885e-01 5.46199262e-01 -1.63103744e-01 -5.88434041e-01 4.21706289e-01 6.07993186e-01 -4.52604502e-01 3.83902639e-01 -1.45731807e-01 -8.03395212e-01 -4.99460429e-01 4.87650722e-01 1.01628065e-01 2.93724149e-01 1.75939761...
[14.36329174041748, 5.125904560089111]
f64680e0-bd68-40ed-bf81-d14cb0db39c0
parkinson-gait-modelling-from-an-anomaly-deep
2301.11418
null
https://arxiv.org/abs/2301.11418v1
https://arxiv.org/pdf/2301.11418v1.pdf
Parkinson gait modelling from an anomaly deep representation
Parkinson's Disease is associated with gait movement disorders, such as postural instability, stiffness, and tremors. Today, some approaches implemented learning representations to quantify kinematic patterns during locomotion, supporting clinical procedures such as diagnosis and treatment planning. These approaches as...
['Fabio Martinez', 'Edgar Rangel']
2023-01-26
null
null
null
null
['video-reconstruction']
['computer-vision']
[ 1.78113252e-01 4.56397504e-01 -4.15985733e-01 -3.83319199e-01 -4.50516671e-01 -9.61987227e-02 3.03397268e-01 -9.29157436e-02 -5.09259701e-01 1.03954947e+00 4.32824045e-01 2.71307558e-01 -3.50900531e-01 -6.06523156e-01 -3.23285133e-01 -7.62506902e-01 -6.94818974e-01 7.98560262e-01 1.70847103e-01 -1.00970373...
[7.1380205154418945, 0.3367210924625397]
1eae87c0-7fa0-4f4d-a982-d8c98575c1b2
constructing-a-multi-hop-qa-dataset-for
2011.01060
null
https://arxiv.org/abs/2011.01060v2
https://arxiv.org/pdf/2011.01060v2.pdf
Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps
A multi-hop question answering (QA) dataset aims to test reasoning and inference skills by requiring a model to read multiple paragraphs to answer a given question. However, current datasets do not provide a complete explanation for the reasoning process from the question to the answer. Further, previous studies reveal...
['Akiko Aizawa', 'Saku Sugawara', 'Anh-Khoa Duong Nguyen', 'Xanh Ho']
2020-11-02
null
https://aclanthology.org/2020.coling-main.580
https://aclanthology.org/2020.coling-main.580.pdf
coling-2020-8
['multi-hop-question-answering']
['knowledge-base']
[-5.18569350e-02 8.53326201e-01 9.72804800e-02 -7.01576710e-01 -1.00367427e+00 -8.05751681e-01 4.92996544e-01 2.19759807e-01 2.26756245e-01 7.61010706e-01 2.87910104e-01 -9.36734200e-01 -4.89645243e-01 -1.21793294e+00 -9.47527230e-01 4.89779860e-01 4.76480454e-01 7.96216011e-01 7.84759283e-01 -5.37305415...
[10.992527961730957, 7.908002853393555]
5887fac9-2784-4066-8562-4623bc2bdede
learning-speaker-representation-with-semi
2110.13653
null
https://arxiv.org/abs/2110.13653v1
https://arxiv.org/pdf/2110.13653v1.pdf
Learning Speaker Representation with Semi-supervised Learning approach for Speaker Profiling
Speaker profiling, which aims to estimate speaker characteristics such as age and height, has a wide range of applications inforensics, recommendation systems, etc. In this work, we propose a semisupervised learning approach to mitigate the issue of low training data for speaker profiling. This is done by utilizing ext...
['Chng Eng Siong', 'Pham Van Tung', 'Shangeth Rajaa']
2021-10-24
null
null
null
null
['age-estimation', 'age-estimation', 'speaker-profiling']
['computer-vision', 'miscellaneous', 'speech']
[ 1.29875541e-01 4.66123611e-01 -1.46042362e-01 -7.43001640e-01 -8.84254813e-01 -1.34062499e-01 6.88729644e-01 3.97519886e-01 -2.78671503e-01 7.02968955e-01 3.52056235e-01 5.64930476e-02 -2.61165828e-01 -4.68307763e-01 -2.45415792e-01 -8.33708644e-01 2.72578862e-03 5.74605644e-01 4.86884937e-02 8.23585242...
[14.213830947875977, 6.097212791442871]
4fb2b67c-c06a-4021-945b-654a69db09af
dual-domain-self-supervised-learning-for
2302.09244
null
https://arxiv.org/abs/2302.09244v1
https://arxiv.org/pdf/2302.09244v1.pdf
Dual-Domain Self-Supervised Learning for Accelerated Non-Cartesian MRI Reconstruction
While enabling accelerated acquisition and improved reconstruction accuracy, current deep MRI reconstruction networks are typically supervised, require fully sampled data, and are limited to Cartesian sampling patterns. These factors limit their practical adoption as fully-sampled MRI is prohibitively time-consuming to...
['Michal Sofka', 'James S. Duncan', 'Chi Liu', 'Kevin Sheth', 'Seyed Sadegh Mohseni Salehi', 'Neel Dey', 'Jo Schlemper', 'Bo Zhou']
2023-02-18
null
null
null
null
['mri-reconstruction']
['computer-vision']
[ 6.21177673e-01 1.35072872e-01 -3.53251338e-01 -6.22726738e-01 -1.11639178e+00 -3.09419394e-01 2.12704867e-01 -1.67046353e-01 -3.54508519e-01 6.91689968e-01 3.17151099e-01 -3.56990010e-01 -3.26435417e-01 -3.65092367e-01 -8.06175768e-01 -7.99993634e-01 -5.02864540e-01 5.67878842e-01 1.55944929e-01 2.20796198...
[13.543734550476074, -2.410111427307129]
da9507eb-2580-4df8-b7c2-083abaee5210
learning-better-representation-for-tables-by
2010.07606
null
https://arxiv.org/abs/2010.07606v3
https://arxiv.org/pdf/2010.07606v3.pdf
Learning Better Representation for Tables by Self-Supervised Tasks
Table-to-text generation aims at automatically generating natural text to help people to conveniently obtain the important information in tables. Although neural models for table-to-text have achieved remarkable progress, some problems still overlooked. The first is that the values recorded in many tables are mostly nu...
['Dayong Hu', 'Linjun Shou', 'Yinliang Yue', 'Can Ma', 'Liang Li']
2020-10-15
null
null
null
null
['table-to-text-generation']
['natural-language-processing']
[ 2.32768446e-01 4.20055777e-01 -5.15319109e-01 -3.21196944e-01 -6.98629916e-01 -3.81253988e-01 5.75499535e-01 5.25168121e-01 -2.55723327e-01 1.29563820e+00 7.65134513e-01 -2.34513074e-01 1.71399862e-01 -1.22111988e+00 -8.38700533e-01 -4.23817307e-01 1.25990778e-01 8.34180892e-01 3.18805158e-01 -8.23585868...
[11.677376747131348, 8.820026397705078]
dc6b8da7-54eb-4c8b-982b-ccfbce354561
revisiting-domain-generalized-stereo-matching
2203.10887
null
https://arxiv.org/abs/2203.10887v1
https://arxiv.org/pdf/2203.10887v1.pdf
Revisiting Domain Generalized Stereo Matching Networks from a Feature Consistency Perspective
Despite recent stereo matching networks achieving impressive performance given sufficient training data, they suffer from domain shifts and generalize poorly to unseen domains. We argue that maintaining feature consistency between matching pixels is a vital factor for promoting the generalization capability of stereo m...
['Edwin R. Hancock', 'Tatsuya Harada', 'Jun Zhou', 'Lin Gu', 'Yimin Chen', 'Lei Huang', 'Chen Wang', 'Xiao Bai', 'Xiang Wang', 'Jiawei Zhang']
2022-03-21
null
http://openaccess.thecvf.com//content/CVPR2022/html/Zhang_Revisiting_Domain_Generalized_Stereo_Matching_Networks_From_a_Feature_Consistency_CVPR_2022_paper.html
http://openaccess.thecvf.com//content/CVPR2022/papers/Zhang_Revisiting_Domain_Generalized_Stereo_Matching_Networks_From_a_Feature_Consistency_CVPR_2022_paper.pdf
cvpr-2022-1
['stereo-matching-1']
['computer-vision']
[ 3.74608099e-01 -1.31729692e-01 -1.30549476e-01 -5.90716660e-01 -4.79387224e-01 -5.25023282e-01 7.51233697e-01 -1.75362766e-01 -4.05260414e-01 7.24068642e-01 2.45883539e-01 8.72865170e-02 1.49295563e-02 -7.49154687e-01 -8.79267037e-01 -7.45626330e-01 3.14952582e-01 1.53358296e-01 6.06362939e-01 -3.07443053...
[8.735492706298828, -2.299800395965576]
9f30915d-9ec2-445a-9b60-acd125931fb5
zegot-zero-shot-segmentation-through-optimal
2301.12171
null
https://arxiv.org/abs/2301.12171v2
https://arxiv.org/pdf/2301.12171v2.pdf
ZegOT: Zero-shot Segmentation Through Optimal Transport of Text Prompts
Recent success of large-scale Contrastive Language-Image Pre-training (CLIP) has led to great promise in zero-shot semantic segmentation by transferring image-text aligned knowledge to pixel-level classification. However, existing methods usually require an additional image encoder or retraining/tuning the CLIP module....
['Jong Chul Ye', 'Yujin Oh', 'Kwanyoung Kim']
2023-01-28
null
null
null
null
['zero-shot-segmentation']
['computer-vision']
[ 7.37217963e-01 1.15678795e-01 -4.26245451e-01 -5.92692912e-01 -1.30950356e+00 -2.95072049e-01 3.86466473e-01 -2.64769495e-01 -5.31345606e-01 2.48435155e-01 -1.68687999e-01 -1.06453687e-01 2.43008524e-01 -5.23393929e-01 -1.13358104e+00 -5.34098506e-01 5.73557794e-01 6.09070957e-01 4.84328032e-01 7.56009221...
[9.7290678024292, 0.8257309198379517]
893b210d-21a1-4545-8a37-3f75b2f9061c
on-the-relation-between-sharpness-aware
2305.05392
null
https://arxiv.org/abs/2305.05392v2
https://arxiv.org/pdf/2305.05392v2.pdf
Sharpness-Aware Minimization Alone can Improve Adversarial Robustness
Sharpness-Aware Minimization (SAM) is an effective method for improving generalization ability by regularizing loss sharpness. In this paper, we explore SAM in the context of adversarial robustness. We find that using only SAM can achieve superior adversarial robustness without sacrificing clean accuracy compared to st...
['Yihao Zhang', 'Jingyu Zhu', 'Zeming Wei']
2023-05-09
null
null
null
null
['mathematical-proofs']
['miscellaneous']
[ 1.29404619e-01 -1.46704791e-02 -9.34248939e-02 -1.53182030e-01 -1.06980836e+00 -1.06615245e+00 3.66605699e-01 -7.63026671e-03 -3.09564590e-01 7.81290472e-01 3.44229639e-01 -5.06901860e-01 -8.91777724e-02 -6.63382530e-01 -9.53029275e-01 -7.28767872e-01 5.52335009e-02 -3.43115211e-01 -9.85088423e-02 -3.20128053...
[5.655416488647461, 7.829474925994873]
60f63592-dda8-4ae8-86aa-21b57e58c22a
brundlefly-at-semeval-2016-task-12-recurrent-1
null
null
https://aclanthology.org/S16-1198
https://aclanthology.org/S16-1198.pdf
Brundlefly at SemEval-2016 Task 12: Recurrent Neural Networks vs. Joint Inference for Clinical Temporal Information Extraction
null
['Jason Fries']
2016-06-01
null
null
null
semeval-2016-6
['temporal-information-extraction']
['natural-language-processing']
[-8.63703638e-02 1.71006292e-01 -6.22772932e-01 -4.08054382e-01 -8.41685571e-03 -9.08429027e-01 6.55310392e-01 -6.53472245e-01 -2.85945535e-01 1.06888819e+00 -4.63127941e-02 -1.01159286e+00 -3.91567826e-01 -9.63214397e-01 -4.95059669e-01 -6.31337762e-01 -9.79754329e-01 7.25764990e-01 3.30370307e-01 -6.93831444...
[-7.221503734588623, 3.6570255756378174]
c278567b-39a0-4f79-b35f-85779da2cb50
robust-multimodal-fusion-for-human-activity
2303.04636
null
https://arxiv.org/abs/2303.04636v1
https://arxiv.org/pdf/2303.04636v1.pdf
Robust Multimodal Fusion for Human Activity Recognition
The proliferation of IoT and mobile devices equipped with heterogeneous sensors has enabled new applications that rely on the fusion of time-series data generated by multiple sensors with different modalities. While there are promising deep neural network architectures for multimodal fusion, their performance falls apa...
['Omid Ardakanian', 'Xin Yang', 'Sanju Xaviar']
2023-03-08
null
null
null
null
['human-activity-recognition', 'human-activity-recognition']
['computer-vision', 'time-series']
[-4.81292494e-02 -3.08614463e-01 3.34018141e-01 -2.67602324e-01 -1.09578383e+00 -1.43254921e-01 6.14204645e-01 2.01769710e-01 -5.74667037e-01 6.99284434e-01 8.63931596e-01 2.42388353e-01 -2.93228626e-02 -5.51765382e-01 -1.02289343e+00 -7.01938987e-01 -1.95732657e-02 5.23458160e-02 -3.78245950e-01 -5.17313540...
[7.638155937194824, 0.8345587849617004]
c082e7c7-d0ce-4c00-8925-d4d8e4f8752a
fast-exploration-and-learning-of-latent
2303.07397
null
https://arxiv.org/abs/2303.07397v3
https://arxiv.org/pdf/2303.07397v3.pdf
Fast exploration and learning of latent graphs with aliased observations
We consider the problem of recovering a latent graph where the observations at each node are \emph{aliased}, and transitions are stochastic. Observations are gathered by an agent traversing the graph. Aliasing means that multiple nodes emit the same observation, so the agent can not know in which node it is located. Th...
['Dileep George', 'Meet Dave', 'Sivaramakrishnan Swaminathan', 'Ishan Deshpande', 'Miguel Lazaro-Gredilla']
2023-03-13
null
null
null
null
['efficient-exploration']
['methodology']
[ 3.94444972e-01 9.21959937e-01 -4.96351302e-01 1.43830806e-01 -8.93792629e-01 -1.01595879e+00 6.13768280e-01 4.42805678e-01 -3.61323714e-01 7.35701978e-01 2.61968911e-01 -3.18246543e-01 -2.12598935e-01 -7.98225522e-01 -6.85694933e-01 -9.09324169e-01 -7.91790783e-01 1.57242393e+00 2.40402505e-01 5.45792639...
[4.308643817901611, 2.1260082721710205]
8e21e55b-eeb5-45d3-b4a1-a2d9ccd9db01
goferbot-a-visual-guided-human-robot
2304.08840
null
https://arxiv.org/abs/2304.08840v2
https://arxiv.org/pdf/2304.08840v2.pdf
GoferBot: A Visual Guided Human-Robot Collaborative Assembly System
The current transformation towards smart manufacturing has led to a growing demand for human-robot collaboration (HRC) in the manufacturing process. Perceiving and understanding the human co-worker's behaviour introduces challenges for collaborative robots to efficiently and effectively perform tasks in unstructured an...
['Robert Mahony', 'Stephen Gould', 'Jiahao Zhang', 'Yizhak Ben-Shabat', 'Zheyu Zhuang']
2023-04-18
null
null
null
null
['action-recognition-in-videos']
['computer-vision']
[ 2.74740607e-01 3.67087126e-01 4.16123033e-01 -4.17630374e-01 7.33719319e-02 -5.94276726e-01 5.80661416e-01 3.02270025e-01 -1.66480869e-01 2.26297036e-01 -1.41426831e-01 4.38141562e-02 -3.23258936e-01 -3.34354103e-01 -3.94954056e-01 -3.16131741e-01 1.04211941e-01 8.05498064e-01 4.83429134e-01 -7.28143215...
[4.907749652862549, 0.7604445219039917]
3962e3ac-447c-4b3f-97e7-9c0b1e0a5418
category-specific-semantic-coherency-learning
null
null
https://dl.acm.org/doi/pdf/10.1145/3394171.3413871
https://dl.acm.org/doi/pdf/10.1145/3394171.3413871
Category-specific Semantic Coherency Learning for Fine-grained Image Recognition
Existing deep learning based weakly supervised fine-grained image recognition (WFGIR) methods usually pick out the discriminative regions from the high-level feature (HLF) maps directly. However, as HLF maps are derived based on spatial aggregation of convolution which is basically a pattern matching process that appl...
['Wanli Ouyang', 'Haojie Li', 'Zhihui Wang', 'Shijie Wang']
2020-10-12
null
null
null
null
['fine-grained-image-recognition']
['computer-vision']
[ 8.38491321e-03 -2.86625713e-01 -2.00669944e-01 -7.83749700e-01 -5.92373312e-01 -6.84434772e-01 6.81740820e-01 -1.02134556e-01 -1.02238856e-01 2.52867669e-01 3.79837215e-01 9.45309084e-03 -4.38880265e-01 -8.00154686e-01 -7.17182159e-01 -9.02020037e-01 4.29007001e-02 3.33622098e-01 2.24305987e-01 -5.28587028...
[9.715191841125488, 2.037099599838257]
0ba6eddd-2b93-4a77-9712-8d9372ff0d60
explaining-rl-decisions-with-trajectories
2305.04073
null
https://arxiv.org/abs/2305.04073v1
https://arxiv.org/pdf/2305.04073v1.pdf
Explaining RL Decisions with Trajectories
Explanation is a key component for the adoption of reinforcement learning (RL) in many real-world decision-making problems. In the literature, the explanation is often provided by saliency attribution to the features of the RL agent's state. In this work, we propose a complementary approach to these explanations, parti...
['Jayakumar Subramanian', 'Georgios Theocharous', 'Chirag Agarwal', 'Nan Jiang', 'Balaji Krishnamurthy', 'Arpan Dasgupta', 'Shripad Vilasrao Deshmukh']
2023-05-06
null
null
null
null
['offline-rl', 'continuous-control']
['playing-games', 'playing-games']
[ 2.12509051e-01 4.39310044e-01 -4.70670730e-01 -1.97056472e-01 -2.95092136e-01 -5.58740437e-01 7.88022459e-01 2.81394571e-01 -4.10174757e-01 1.01519454e+00 3.28116119e-01 -3.58865321e-01 -2.30783418e-01 -4.71000046e-01 -9.38276112e-01 -5.15390933e-01 -2.90052176e-01 5.28160334e-01 9.81160030e-02 -2.88455248...
[4.188821315765381, 1.701714038848877]
6ce07795-6e9f-4182-b218-219f261c723b
the-limitations-of-large-width-in-neural
2106.06529
null
https://arxiv.org/abs/2106.06529v2
https://arxiv.org/pdf/2106.06529v2.pdf
The Limitations of Large Width in Neural Networks: A Deep Gaussian Process Perspective
Large width limits have been a recent focus of deep learning research: modulo computational practicalities, do wider networks outperform narrower ones? Answering this question has been challenging, as conventional networks gain representational power with width, potentially masking any negative effects. Our analysis in...
['John P. Cunningham', 'Geoff Pleiss']
2021-06-11
null
http://proceedings.neurips.cc/paper/2021/hash/1b9f38268c50805669fd8caf8f3cc84a-Abstract.html
http://proceedings.neurips.cc/paper/2021/file/1b9f38268c50805669fd8caf8f3cc84a-Paper.pdf
neurips-2021-12
['l2-regularization']
['methodology']
[ 6.06155284e-02 4.69369709e-01 -1.18982047e-01 -1.06969262e-02 -3.77430499e-01 -7.14005709e-01 5.69222033e-01 1.29982056e-02 -5.22088349e-01 7.09815204e-01 1.97148532e-01 -5.54042578e-01 -3.98667663e-01 -8.77039194e-01 -9.10708666e-01 -1.15739346e+00 -1.92487001e-01 4.70463514e-01 4.59274828e-01 1.42772332...
[7.820926666259766, 3.6821727752685547]
286e8258-a7b9-4b73-97b0-1e0a326a4fe3
matt-a-multiple-instance-attention-mechanism
2209.04109
null
https://arxiv.org/abs/2209.04109v1
https://arxiv.org/pdf/2209.04109v1.pdf
MATT: A Multiple-instance Attention Mechanism for Long-tail Music Genre Classification
Imbalanced music genre classification is a crucial task in the Music Information Retrieval (MIR) field for identifying the long-tail, data-poor genre based on the related music audio segments, which is very prevalent in real-world scenarios. Most of the existing models are designed for class-balanced music datasets, re...
['Menghua Zhang', 'Xiaokai Liu']
2022-09-09
null
null
null
null
['genre-classification', 'music-information-retrieval']
['computer-vision', 'music']
[ 1.48319885e-01 -5.70496380e-01 -4.37535912e-01 -1.30174294e-01 -1.32658005e+00 -4.70007032e-01 8.72564875e-03 1.11686006e-01 -4.23326008e-02 2.58832425e-01 4.66160029e-01 2.09790856e-01 -5.16759098e-01 -5.42990506e-01 -5.73329151e-01 -7.20111787e-01 7.37395212e-02 6.04740977e-01 -3.05119038e-01 -1.56294823...
[15.759607315063477, 5.2430572509765625]