text stringlengths 54 548k | label stringclasses 4
values | id_ stringlengths 32 32 |
|---|---|---|
The digamma functions appearing in the expressions are easily calculated using the formulas {{cite:01f520c49f2b0978ccd82e3072c4bfe2c5d66268}}:
{{formula:76f44841-ec47-4953-8d13-a6e5fea7d954}}
| r | b37e7f6bdd98420e5941381db7428156 |
In the recent years, pre-training of deep neural networks has emerged as an effective approach to overcome the problem of data scarcity {{cite:2da76d4742990e30b10d207438db4f9bea47e5e3}}. The key idea of such a technique, which is also called “transfer learning", is to learn general representations in a setup where subs... | m | 5be087444ed503e2926d5827604e14fc |
Baseline Methods. We study the effectiveness of our problem formulation by blackbenchmarking against two language-conditioned baselines: Image-BC and C2FARM-BC.
blackImage-BC is an image-to-action agent similar to BC-Z {{cite:848b5cb8c85f5d3328e57b9ac19ab03130cbe753}}. Following BC-Z, blackwe use FiLM {{cite:3110825b14... | r | 68930f45d169d3a91a1cfd8724c7bd36 |
During the last years, non-linear equations involving general
integro-differential operators, especially, fractional Laplacian,
have been studied by many authors.
Caffarelli and Silvestre
{{cite:d2c86ac8591252cb9e3087f97527466c91454323}}
gave a formulation of the fractional Laplacian through
Dirichlet-Neumann maps. Var... | i | 9249cd57655c253aa413abade98aa7ed |
The objective of self-supervised anatomical embedding (SAM) is to encode the semantic anatomical information of each pixel, so that similar body parts in different images can have similar embeddings. Inspired by previous works on contrastive learning {{cite:790869e5b4e4f850e0fdf823fd08576cceb518f8}}, {{cite:6ac8652f808... | m | 56b2e7278e14836bebb6e006c2754595 |
In our model, if the receptive fields of two latent representations do not overlap, they are naturally disentangled. For high-level representations, we propose to utilize a sparse prior to encourage disentanglement. We find that if the dataset only contains a few high-level factors, such as the 3D Chair dataset {{cite:... | d | 20292401c63bee74ce8748a814b66dba |
Short-term forecasting up to twelve hours in advance allows for predicting weather conditions with higher spatial and temporal precision than longer time ranges. This makes it possible for these forecasts to have substantial impact on society by helping with daily planning, energy management, transportation and the mit... | i | d06ef93009b5e2a427e462456673432f |
The topological diagrammatic representations and our results of branching fractions in the {{formula:05c5c8bf-4b6f-40a0-a140-1073c6d3685b}} and {{formula:540fa2e5-cde8-4ddf-86c6-4443c8a90a80}} decays are presented in Tables REF and REF for the {{formula:27ee2dba-f3e0-4a8e-a829-c63721451f8f}} and {{formula:bcdb2c07... | r | c9afbdf1c147811d4a7031e7c30aa8d6 |
Arbitrary Face Editing.
In real-world scenarios, a fine-grained and wider-range control for each attribute is very useful. Previous works such as RelGAN {{cite:793c37f522db2291433c1fa74e22c18d6a886808}} and InterFaceGAN {{cite:c516545025a5c390b982ab6c9f0d86c191f0afa2}} also provide continuous control for facial attribu... | m | 218e18ffc8ab3fe6d83cf9acc8d1bb83 |
For the ease of illustration, we consider the case of two non-overlapping subregions (see fig-domain-decomposition) in what follows, where the interface conditions are invariably of the Robin type {{cite:39b967cea59eb4c0ef82203e4522e23995acd5a6}}, {{cite:f21a7c2ed8ea745129a80222c27501b2406de4c3}}, {{cite:769f49b0034268... | m | cd4bfa9b3f677f16f1343cbe0f5bf57b |
In order to validate the generalizability of the EL module, we apply it to deeper networks, such as ResNet50 {{cite:b610170c501930e633ed058210a481007f54f568}}, which are rarely reported in existing literature. The results in Table REF shows that EL-ResNet50 (65.6% top-1 accuracy) is superior to the Bi-Real-ResNet50 (6... | r | bb25de8624a2dae6c77a87808fd9655c |
Let {{formula:beec9933-b683-4303-9cf3-305f000796e8}} be an eigenvector corresponding to
{{formula:9b490ae5-9244-4547-a223-8ab006613991}} , i.e., {{formula:4b208bee-89ef-46b3-8303-ab44f0a1c2cd}} .
By the celebrated Perron–Frobenius theorem (see {{cite:413d5844c6240e700e8bcce00f2e36221aee291c}}
or {{cite:73c0ef0bb62e6e4... | r | c67030cf07d4da10b7034beeb3eb4f0a |
Uncertainty Sampling: Least Confidence (LC) {{cite:adc33a0d5b94f220985007b8ee202ac27a78db2f}}, Max-Entropy (ME) {{cite:6f49abd39b0fe5620201130b9d4e0f0b499c3354}}, Min-Margin (MM) {{cite:11b5e2306234fa8160859644c07d4bf2e57fe8de}} & Deep Bayesian AL (DBAL) {{cite:f13b992ebe0a7dd8e7848aeff4f7c4e5e425ec7b}}
Diversity Sam... | m | c2e9460673856dff25fea6b36c7a9cda |
Our experiment design is shown in Fig. REF including datasets (A and B), the methods development (C) and their assessment (D).
We propose a single-pipeline multi-reconstruction approach as data augmentation for fetal brain MRI segmentation.
To strengthen the generalization of our findings, we assess our multi-reconst... | m | a81455d66b03c801eaa0f0a7dea789a1 |
(2). CSI: It is a recently proposed OOD detection technique {{cite:d898cb2bf0c5f4c883220ccfb5835b27a07649e2}} that is highly effective. It is based on data and class label augmentation and supervised contrastive learning {{cite:a7e1af3752e2f80b72e298b584bee3f5b6e9ddcf}}. Its rotation data augmentations create distribut... | m | 5aa6a3694507f9e35899cb6a459d3f14 |
In the SME EFT formalism {{cite:9b897ddd2229dbde9c3fdf1e559ee3484ad3552f}}, {{cite:922c4b5fa7c623d9f711b7ff7d9ba075e28e24cb}}, {{cite:6b4374af81c7209518d5451b1243208eec140ad3}}, if the apparent cutoff above {{formula:e259acd9-16be-4cb1-9177-041580f7b5d2}} PeV in the neutrino spectrum shown in Figure REF is caused by ... | r | c3485cc972d1e2e1fbb06d364e4db6a9 |
There are many potentially relevant sources of information beyond an agent's own experience, or that of other humans or agents. For example, it has been shown that YouTube videos are useful for learning to play the Atari game Montezuma's revenge {{cite:97aae43f95f5c9b275f99c1df1f1b62696af915d}}. Training an embedding n... | d | 1db3680a7bcc9cb905a0a917b59210fe |
It is known that multiple instances in the single natural image possess the co-occurrence relationship, and usually have different semantic meanings. Therefore, the models should have the ability to distinguish the semantics of different instances. However, it is still challenging to discriminate different instances re... | i | d7f150160435f6f77fcc23499a9bcb28 |
Text image:
Images containing text usually appear in images obtained by scanning a text.
In recent years, with the expansion of the use of smart phones,
scanning software has expanded {{cite:bd6172092011bce0ce05e26e15a0c84b59268e35}}. Various factors can affect the quality
of the output image, such as hand shake when u... | r | 9eb13010bc0edec039f3b03b13b79bb3 |
From reactions fueling cells in our bodies to internet links binding the World Wide Web into a small-world structure, networks are at the basis of many phenomena.
Random Graph theory sets up a common toolbox studying networks independently of their context from a probabilistic point of view. The most basic random graph... | i | 90f8cca71f2109057ff4b55b8b75179b |
To calculate the anomaly score of a sample from the Softmax probabilities, Golan et al. (2018) {{cite:f4453f0bb11f90a1f3c7f27654601313351dbd9b}} combined the log-likelihood of the conditional probability of each of the applied transformations:
{{formula:6a63a15d-93cd-4b18-abfe-ccad1918f5dc}}
| m | 5fe9c1c6e74fe49f44e6598292411ed6 |
Regarding the spontaneous microphase separated regime, this is notable especially because there is no term in the free energy density {{formula:325ab20d-d7ba-4c52-aa7d-b4bfea9f4ba9}} which explicitly favours demixing. In other words, in the absence of {{formula:a2464972-98ce-436f-985b-d2b25d63c11a}} the system would ... | d | a38585172548f73ef1de998a546fb019 |
It is difficult to obtain all the structural
data from the above pool of published articles.
It is rather practical to use chemical compositions
as the direct descriptor.
In the preceding studies, {{cite:200c5ea1a4d9ae94ca793301f684650deebc30ee}}, {{cite:038112fa81d29e31b94e4708a82d34969a6a84a1}}, {{cite:ba75f77aad6e1... | m | eb2fd211984cc267919e2903bda970ad |
Last but not the least, as a physical model at cluster scales, the {{formula:3dfc89b1-97f2-4f04-bebe-536f2a705e0b}} should be subject to more observational tests, just like the NFW profile of the dark matter {{cite:e552d46e19b09d23e157e3cf32ac947f8fd16948}}, {{cite:58397792d88e469d4a34ff41d46634bca26045cf}}. A new cha... | d | ae600da969837b1a3267b602d701203b |
Finally, we see similar ideas circulating in the neuroscience community. A recent neuroscience commentary, “What artificial neural networks can learn from animal brains” {{cite:572f0afd9c9a9a737cd6e4db897bde99155342c7}} provides a critique of how learning (and also meta-learning) is currently implemented in artificial ... | d | 923435796067434909e7422b7d2f0f77 |
Effectiveness of tracker hijacking attack. Fig. REF shows the average number of frames that the AEs on object detection need to fool for a successful track hijacking over the 20 video clips in the evaluation. Although a configuration with {{formula:a9a080ce-6ab3-4208-aed3-b6e786ecd4b5}} and {{formula:0e513525-cf69-4b... | r | 1a170f7ad550aefb46ea8fa02cb690a3 |
where {{formula:df4ad3a2-5563-4757-8435-32c7d9e3368f}} is the graph cut of a weighted graph with {{formula:f2d3ba0e-c31b-4777-8846-3b3696465d50}} vertices and the adjacency matrix {{formula:9ecc01fc-e4c6-4e46-aae2-1f5c923e1d8b}} .
Hence, the balanced graph bisection problem {{formula:412000fe-2f04-490a-bfbe-bf72717d8... | m | 785ba0ae292b3aedf86e5473e1298876 |
Today, a large number of text and written materials are present in English. However, with roughly around 6,500 languages in the worldhttps://blog.busuu.com/most-spoken-languages-in-the-world/ {{cite:a64edbb9178ce4b2d81c4ebe39774d2f052ae807}}, {{cite:c92e08d29155fe7fde78eed422ad332d839e9565}}, {{cite:c4bdc940b3c0a71311d... | i | 26b17225436cc2856d3dbea88cd273ac |
The Hamiltonian Eq. (REF ) [Eq.(1) in the main text] is
essentially the same as that used in
Ref. {{cite:c5152e5125d8b482173dc14d789f6c5a4cb251ab}}, {{cite:836e604ab210138eea65c904a9ceec0af111e6c9}}, {{cite:1f2fed818937ced879770dce57371a8327d99e4a}}, {{cite:430009c2aa0cd3e7381f22a12a0b4a1075a5b394}} resulting from the
... | d | 0296c776e2093e67ecf5049f87f624bc |
which can be checked by plotting the autocorrelation function in lin-log scale as shown in Fig. REF . Here, {{formula:7b24f17c-5997-4b0d-9dc4-82470c0774c2}} is the exponential correlation time.
The correlation time gets higher with increasing {{formula:26f67423-4fbc-4e43-8a57-6bef37c49c6c}} . The curves show that ther... | r | 87d03c36678cf20b98f464cc7e371b6f |
To further improve the recall, one possible direction is to learn a joint model for detection, tracking, and prediction. The intuition is that, future predictions can be used to infer the subject's location during tracking, whereas accurate tracking results can improve the prediction accuracy. Recently, several work pr... | d | b372fb50ae6d4e5b33b05e5de3893f4c |
Although attribution methods have become an active and important area of research, their general applicability and usefulness have been scrutinised and criticised. For instance, Rudin {{cite:ad85a47942528c7ccb4fbf200b9e4d17d9fbcc9a}} argues that explanation and, in particular, attribution methods cannot be completely f... | m | 3bb2c3ef68d6ea675d15d2f0a8e188d1 |
Theorem REF and REF apply directly to models of random maps when all faces have the same degree to deduce the following corollary which extends the aforementioned results of Le Gall {{cite:776bc19f1b2d8c8530abaa9c90787aa76f34bcf2}} and Bettinelli & Miermont {{cite:e62db669ddf12b30d02c7ecc81d174e79c090927}} when the s... | r | 05ede55fa80ab3e8af62a0f2decab7ef |
Table REF reports the evaluation metrics of different normalization layers on the Split CIFAR-100 and Split Mini IMN benchmarks. We consider the averaged accuracy at the end of training ACC({{formula:a77780bb-9ab4-40c9-aef8-ecc01850f197}} ) {{cite:f4dcc3eb7b395f828433e1c652b8393e4f47c1fc}}, forgetting measure FM({{fo... | r | f66384542191efdbbfe6bd0952a04749 |
Let {{formula:832ced75-320e-4633-9975-f2112db06da3}} be constructed as in Definition REF .
We may write {{formula:ee2c49f4-9b9f-4952-8603-c5560f237ccc}} , by symmetry and independence of {{formula:8df8a57e-f3bc-4f8b-87fd-598f95348239}} and {{formula:650a4824-2283-4763-b71b-b1a157458077}} (c.f. {{cite:2bf586ca0536b00... | m | bba8b516630a3825971baf9525212df2 |
Beyond first-order methods, the most developed methods are maybe second-order methods, among which the closest to our setting are cubic-regularized Newton methods originating from {{cite:61e8574e3e4c3d9965e5c97dd0f9d078f22d39ff}}. This line of works includes the development of accelerated methods {{cite:1652cb318fe8b4d... | m | 4452f3bc8f408530d87f284fcc9c2815 |
Since the aforementioned problems are intractable, relaxations of these problems have
been considered.
The problem of computing chordal completion that are inclusion-wise minimal,
has been intensively studied, either to find an optimal tree decomposition with an exponential
time algorithm or to find one tree decomposit... | i | 121d45036d84296287821ec942610d0b |
With the rapid development of the World Wide Web, most of the human activities are migrated to the Internet,
such as game {{cite:826ea83411af48f51bd796e1f865703a25fe3c26}} and social {{cite:fccc536f00815f4b10ae2e4b216d2711ca4ae757}}. Online service providers often develop efficient toolboxes to improve our online exper... | i | 0eaa337c4e631c4c82e2f6eda5d61af0 |
These errors are intrinsic to the data and current mapping algorithms. As seen in Table
REF when using three different popular Burrows-Wheeler mapping programs
for Illumina data on sets of simulated chicken read data{{cite:d83a70372c7a9640501a2a9044ac6e227655d485}}, {{cite:744814896a7600a214cf2daf5ceaf502ec7adfb5}}, {... | r | 668cfddf611c6ecac334b159b41e67f4 |
The authors made use of the fastText C++ library (with default hyper-parameters, except where mentioned) by {{cite:5d2cd79d3b2637307154769ed4a974ad31590688}} to generate 8 word2vec models and 8 subword models from each corpus, based on the optimal hyper-parameter combinations demonstrated by {{cite:6a7fdfc2d49f69c2ec47... | m | 07e1787feadca667b5c9364e14fe7c22 |
G+0.693 is known to be one of the most chemically-rich interstellar sources in our Galaxy {{cite:42675d632264eab9cf447af19c3a6cf153606c4b}}, {{cite:40efa402531affd063678e487741985e0bf96c8e}}, {{cite:dabd6ecd94b1d20128c47cd29c45992a056445b4}}, despite the fact it does not show any sign of high-mass star-formation activi... | i | a0d038abd7373bf480c42bddcd990e01 |
The number {{formula:dcc61581-d94c-49f7-b9dc-a461ea126d1c}} is the number of eigenpoints of a general tensor in {{formula:80ac4b82-9ead-4958-b45a-bdb7ca8a908a}} , as proven in {{cite:c58286be2f267608432002181ca35fd8c9541a64}}.
| r | 6e471cf237c8b59f7f618cfc20964acc |
Lundberg and Lee {{cite:7ada4801ca775943f515f11ad1bbe0823094bd28}} propose a model-agnostic kernel approximation of Shapley regression values described above. They also demonstrate that both LIME {{cite:37447985b29b72001c0b9b1909993c373bb6cc9b}} and DeepLIFT {{cite:89a9a6a825be6ec7fc8d3b9d3fa25dcbc328e1f8}} are special... | m | 5239336e14b52ed85518be3dee803608 |
The fact that the condition {{formula:9cd8a84f-dc7f-473e-9ee2-5dbd5e5b3b39}} holds for curves with involution is known. In the CKP theory {{cite:31edd2b5219aace10e1f2e9744ad9e1f3b1fc0db}}, {{cite:288913801967f6b6e1adc3ceac9f9d851d959374}} equation (REF ) plays the same role as equation (REF ) in the KP theory. Namely,... | i | a3e6f2c4fb7182296f93cf7181cb2c4a |
Change in the polarization of laser field also affects the dynamic phase of electrons undergoing intra-band motion, ultimately affecting the excitation rate during multi–photon absorption.
For example, circularly polarized lasers are also important as an
ultrafast laser waveguides{{cite:9c724244b847cee88f4cce0affdada38... | i | e4e37d0b5b53cd4c15af4f5a77b171ce |
To verify the effectiveness of our proposed model, we conduct our experiments on three publicly available datasets for NER. These are CoNLL 2003 {{cite:60a9af5f62f33c44ffc433597a3d4fe3c241614f}}, CoNLL++ {{cite:3684e63a8140d011f5f0577216807b633b7e2c7f}} and OntoNotes v5 {{cite:86716a1574c3f6f26c06d2aca84efe3ca54c7c00}}... | i | da2128ec96cdf94f14019613815a2224 |
To validate our study, we have compared our results with others in Fig. REF . It shows the dependence of {{formula:c2fae165-91d8-40c4-ae85-000ebc123159}} on system temperature for a fixed system size at zero chemical potential ({{formula:ecae0157-b207-451a-b1cb-418edd2a6a44}} GeV). The AdS/CFT lower limit to {{formul... | r | d4ec6628e8571fd156460360b2d5fe4d |
Convolutional neural networks (CNNs) have achieved state-of-the-art results on real-world applications such as image classification {{cite:7cdfe60c5732e5c8465ebc25a1b19a3b187a1926}} and object detection {{cite:3d7f0ce1f61d88087d294b0e5ed54d665f0f63ee}}, with the best results obtained with large models and sufficient co... | i | 400615ba09834279dd2b7f424b0fb30d |
For the existence of polynomial solutions of Heun equation, one of the parameters {{formula:eb8ebae8-e018-40ab-ac0e-90cda96e1ceb}} or {{formula:7ac2560d-0adb-4fdd-abfd-4d0e880ed40d}} must be chosen conveniently. For instance, for the class VIII Heun polynomial these parameters are chosen to be {{formula:e2f18e39-954b... | m | d3de5b2bc9fca3e0937e323488340f88 |
Moreover, we test state-of-the-art deep clustering methods that learn deep representations and cluster assignments jointly. It has been studied extensively in recent years {{cite:8d2fe2da5846e7e6d53e091d2ba6917bf3e529ac}}, {{cite:db3471b62019661c9740359da4b453e06ad3659e}}, {{cite:63cae24c0880056a821276075095c8d7c5d8287... | m | 4ad5681687508fb3bd5ee971d0baf7ef |
Since {{formula:f71dd79a-6d9a-44e1-95c8-932606ed3c59}} is {{formula:e37dc2d7-41ce-4bd4-bde6-21e075a78e62}} -convex for supports, Theorem REF (a) implies that {{formula:2de04d7d-6ba1-4b9d-88c3-af4532bfcf7c}} satisfies the minimum principle in {{formula:30acc5d3-adff-4a24-9e8f-61b7e9e2ce59}} for every {{formula:eb958b... | r | 5eea0cee287a143be6e27cb0e58ecb5e |
In order to study networked systems properties, it is possible to exploit a method allowing the creation of a random network by using a given degree sequence. To this end, one of the most common and used tool is the configuration model {{cite:a4dad8ee2321925714c65c66b60c06bcb1555778}}. This latter, defined as a general... | m | 6de51a4b20cb209b8bd75a23fdf06318 |
The class is defined for each {{formula:e6dd65b2-902b-4264-b965-74dff94a7c30}} here we shall focus on the case where {{formula:167d33cf-c234-41e4-a5dd-9b18786d7c1a}} which may be generated by an {{formula:8cd504d2-8633-4786-9881-7fcf984f9556}} -stable subordinator in relation to the work {{cite:759d73a219457c510e936d... | i | 8391cd30f48bd556de8d8aeed45fddd6 |
In {{cite:0d2ac08305a9478ebfcace0e38ccfedbe7c1bd52}}, the above bound was developed for the DA setting. We have adapted it to the DG setting by identifying “source” domains in DA with “seen” domains in DG and the “target” domain in DA with the “unseen” domain in DG.
While the first two terms of (REF ) and (REF ) are qu... | r | 7c11320e2523bb45c205dfdcfb7808d9 |
ii) to observe resonance {{formula:de1e7303-b1b9-4065-a5af-b14e7f3c2bcb}} in {{formula:79266f90-59ae-418a-ba10-5642f4525fa6}} {{cite:7b388abdf1f3c6dc577fbfcc8ee29644326c2be1}}. In {{cite:be3dc9b7f0bc2103f39041a1dc54724c4f5580e6}} the data fitting
indicates that the enhancement is a S-wave Breit-Wigner resonance
{{for... | d | 5393ac52c9780ddc6a3c099bf723d106 |
Future Link Prediction: We report the tranductive test AP scores for future link prediction in Table REF . Unlike previous work which focuses on {{formula:31d42dbb-b629-43be-8c05-1f23a60e8b23}} , we evaluate all models under {{formula:0709738b-ea0a-418f-8bd4-4d2023f9818a}} to test their medium- and longer-term forecas... | r | 81395294c5f47ae9341da2e9e8978512 |
Recent studies {{cite:7b7bbce3424d111bdcc1e3cad40a23c538110718}}, {{cite:01f6cd4d4da207b1bef1939575fe3222955eefdd}} have proven that despite different propagation processes of various GNNs, they usually can be fundamentally unified as an optimization objective containing a feature fitting term {{formula:037ab3f0-6e31-4... | i | 1c50a1eab6e8de0aea3e5694ebc0ec24 |
For resource unconstrained PDA, LoCO-PDA is compared against ETN {{cite:4b0496cc928fc3256c05161a06207d5b6d1d61f2}}, IWAN
{{cite:f7531315c5a198edd095aa04881237f1b069c744}} and PADA {{cite:439155183a4f446343ce907c8d0b4c6a456c4886}}.
| m | af6c8aeb6c04cdee0a63197956ff7b65 |
Recently, we can easily obtain high-frequency data such as stock price data and life-log data (blood pressure and EEG, etc.) thanks to the development of measuring devices,
and statistical inference for stochastic differential equations based on high-frequency data
has been developed. For parametric estimation of diffu... | i | 7d3229bab14db834c960a69a98c50b63 |
General relativity is the most successful theory of gravity, which could explain a variety of observations, including the bending of light, perihelion shift of Mercury and the recent detection of the gravitational waves.
Despite this great success, there remain some issues with the general relativity which are not yet ... | i | 5fb5224ef58dcd0ac95971695b7ae35f |
This performance difference suggested the possibility of a combined, fused classifier of our proposed relational classifier and the more common CNN based low-level feature work of {{cite:07ae01524cce46846388400f5d26db80bae09247}}. We combine our model with the predictions from the baseline CNN, the pre-threshold proba... | r | dbd7142c3cdad1c52df1a75de5a772e6 |
Our results may depend on the model setup, in which new blood vessels are created as endothelial tip cells move from left to right. In the future, we may explore the use of extended persistence {{cite:f19e97181b1d189a18aa6faeaa084cfead1b5c6f}} and/or PH transform {{cite:df310d088b240f555ee46a78722e23535b58d828}}, which... | d | f9e6e2ef8946ba84f20b8750b9ea4296 |
Fine-tuning is the dominant technique for using pre-trained vision models to perform downstream tasks, which has contributed towards significant advances in a wide variety of computer vision problems {{cite:7b85e0aee43d97c8b17c57e0fcb7db199fefcd3a}}, {{cite:6f3b89cae56a8deb1c9e47d939ec87ba7dd0b0d1}}, {{cite:7c3f9212985... | i | a2159ecaeb01c9851bf25b269a493a05 |
Second, it is expected the topological phase transition is second
order {{cite:b973b14934d7180dd60bbb8398d3412e85591b0f}}, {{cite:f7bd2d27828ef38fa5260cfea328115c89b9a0d6}}, {{cite:ea9293b47c6eb99ee4db46a1814dfaeb1e05fe75}}, {{cite:231b640253b202ae40c1bb9799b7cc74509f6e8a}}, {{cite:d630faf9c7262e52e93d1f2be37f77f7b69e9... | d | f24847671bffc3e4162943fe38ee4a31 |
An important test for theories of star formation is that the successful theory needs to reproduce the fraction of binaries in dependency of the mass of the primary stars. Most observations have shown that the fraction of binaries decreases as the mass of the primary star decreases. This appears to be in contradiction t... | i | f7bbd3b14cab17a26a88b6bd55e441fd |
Hate speech {{cite:e4fb9ca19140d8c453763182019ce01ae563610f}}, offensive language {{cite:73aa876371c886fb474c9004e0b9d81f5ba52198}}, {{cite:6d24a5e7ef041ec8b73ad2596fb5de1046e38fb2}},
abusive language {{cite:e55d764ee358cfeaf13029d3e5990a6f3d742c08}}, propaganda {{cite:de96e8aacb31694e0702aec76fcb5284d4f121a9}}, cyberb... | i | a612a6963b16990e58228b5297638b53 |
We conducted extensive experiments on CIFAR10, Speech Commands {{cite:2f2fbb866cf7db5b4d4f37d49795fe280939f62f}} and FEMNIST {{cite:8d1e4c9dbe8adc66af7e471a99d51f8b573df360}} datasets. The CIFAR10 experiments follow the same experimental protocol as for baselines experiments. The three local sparsification strategies a... | r | b30d4b46e6ba53e5c5a28d8aefbca67d |
Next we will explore the relative contribution of low- and high-frequency phonons to the heat released into the system through the contact region. In Fig. REF we plot the phonon spectra {{formula:c084a7f4-9068-44da-b120-bbb13dc3174c}} of the interface oscillators at the left and right sides of the contact for both th... | r | fa476af4ca5484b66e781996beae7a75 |
Several directions have been taken to understand these phenomena. The interest in the stability of SGD is one such direction{{cite:1cb95d4f6671046914cd3c80b93baab9e0bff9d6}} {{cite:b32444282e462a346976462b5720956be3ce61d0}}. Others have proven that gradient descent can find the global minima of the loss functions in ov... | i | 7e1dcc15f8667527ea248b0d37130025 |
We see that our approach provides competitive performance, both at 100 % of the labels, as well as there is only a small drop when reducing supervision by factor 20{{formula:c2f99fd1-9900-44b8-9340-a97780d12146}} .
Our mAP at 100 % of the labels is better than both variants (with and without color) of 3DSIS {{cite:7496... | r | 711a07ce7275f46e31e04d3b60635194 |
The argument by Misner, Thorne, and Wheeler {{cite:4ee4f13e3af8d0161a5ef7503ac267e029e7f895}} begins with
what the authors stress is a definition
{{formula:00ee1b02-6961-4dd6-96b7-7d1fabc7f59b}}
| d | d1d0caced36e5f49733e439a8fe853a5 |
The variation of the in-plane London penetration depth, {{formula:113aa005-19b8-451f-b742-fe9bf544dfef}} ,
was measured using a sensitive self-oscillating tunnel-diode resonator
(TDR) described in detail elsewhere {{cite:eaa999c610a743af2e9c0f8709ec681730c84a65}}, {{cite:f24888a90d5f324de056d0fd4be258bb4cd7f852}}, {{ci... | m | 13c64d6d0b11df0946210d5c46ea7db7 |
Generative latent variable models have great utility in scientific and clinical trial analysis to improve hypothesis generation and experimental design {{cite:b9bdab54613f6f6093251828191f5b589dea0e29}}, {{cite:aae054476c6f2ebfcc7571c64e7521a07dcda82b}}. This scientific and clinical utility often depends on two goals, o... | d | 16c44d48858859dee7a14f0086effce5 |
These are all the minimizers of the Sobolev inequality, up to scaling. There are many interests regarding to the stability of (REF ). From the perspective of discrepancy in the Sobolev inequality, {{cite:1e59a8f02a3bffe5ad225c0c0dca278f620aeeed}} gave a quantitative estimate near the minimizers, that is
{{formula:5dfe... | r | febf6a559c1a2f6ee149e921b4cac797 |
With Figure REF , we have demonstrated the overall superiority of our proposals with respect to a common penalty framework in group-wise recovering of sparse precision matrices. Nevertheless, when it comes to performing the analysis a single value of {{formula:cf3d5c3c-7a97-4ac4-8e49-f075afd80aae}} must be chosen. Pro... | r | 98ddbbf62b1ebc24aafeca4b0d67e15d |
Overview.
The architecture of MVSFormer is overviewed in Fig. REF , as the seminal MVS pipeline in {{cite:f8399a656085914c5fda56484e39bb92b4e436d4}}.
Given a group of {{formula:1a38f012-0311-4e77-bf2c-0ffa5b86493e}} images with different views which contain reference image {{formula:ca905d3f-6030-4a02-a0ac-25ef98a1cd4... | m | c587c1b46f39aa600748a68e3c3abab6 |
We remark that in the convex setting, theoretical guarantees for large-scale second-order algorithms
have been established (e.g.,{{cite:dacc682bdc01e38f568f38a4e969fcc3e2384558}}, {{cite:9b45ea59c27a021db978a920a89f9d73e12c454e}}, {{cite:ee3046b3763edef0600f0891eb952541fbb6a42f}}, {{cite:3d893fb0bac53f4b63d14a078ab2746... | m | a02ee001b74848ee2c5bfa4777a9e1f4 |
blueExisting work in control systems engineering typically restricts perturbations to a subset of nodes, typically the control nodes of a network {{cite:97999d3b7cc13c8f2cc7702648bd43b04a4bc8b7}}, {{cite:093195f0805dcf8f3f889e32889a3f177ea09775}} or nodes that translate biologically, e.g., pirin and WNT5A driven induct... | i | de96d60968cc07ac7cbf8a7c4deaffc5 |
Overlap-aware resegmentation. While our segmentation model was found to be useful for both voice activity detection and overlapped speech detection, post-processing the output of existing speaker diarization pipelines is where it really shines. Table REF summarizes the resegmentation experiments performed on top of t... | r | bbd2f8cd05dda75e24397ec535a89019 |
In general, we observe that the framework is sensitive to hyper-parameter changes, and that the various settings of each block affect each other, thus requiring extensive experimentation for appropriate tuning.
From Secs. REF and REF we observe that overlapping patches when drawing positives is detrimental, and light... | d | 49c9419ad1eb9511a26da5358166c964 |
If one wishes to treat general functions {{formula:b009ce46-9ea1-41a8-9a20-6b4c2cfc4d5d}} in the Sobolev space {{formula:a710db34-af84-495f-9758-329ca51e9d48}} , it is necessary to restrict the dimension of the ambient space to {{formula:6b490ecd-12d3-4c52-bc2a-0d5b324c2a51}} , since the norm on the left-hand side of ... | i | 080476237b7ce7fb7792f926e327573c |
In this paper, we propose to apply Bayesian methods based on GP to the inverse problems with sptiotemporal data to account for the space-time inter-dependence. This leads to fitting the whole trajectories of the observed data, rather than their statistical summaries, with elaborated models. More specifically, we use th... | i | 6a82b176e2267a8c5921bbdeb2b20efa |
A further contribution of the paper concerns the evaluation of VaR and ES in the context of portfolio optimization (see, e.g, {{cite:6c43627975c2f287d7316563e05472154fac6f95}} and {{cite:c97af569e38f472f42fef960be3fea31745c230e}}). blackIn recent years, the MAL density has attracted wide attention in the literature for... | i | 8ab69e180a99d1dcff90a68da8746193 |
Approaching clinical presence as a multi-task problem can be understood as an embedding regularisation {{cite:b7fcbe0103c51846dd2066511b9e1fb363ef7b27}} which results in a more clinically relevant representation for survival prediction. However, despite multi-task improvements observed in the literature, theoretical fo... | d | 5d6de84689f471889da2f1c36fa8729d |
Existing work on explaining GNN predictions can be categorized mainly in two directions: 1) factual reasoning {{cite:1a7e601718092d900b0a5415429f6ba5d70566f7}}, {{cite:2fbfe15ebab8ef336ebb3612d2161f4064bf93ca}}, {{cite:c21954a357ce98c8eb9cec6adcd983cba24723eb}}, {{cite:994e55cd76dbcb3dddefeee25db0967c68a519b7}}, and 2)... | i | a947f861e38fd796a594b5dc58ecbc67 |
Probabilistic models allow us to make quantitative inferences about the behaviour
of complex systems, and are an important tool to guide their use and design. When such models are learnt from data, exposed to potential distribution shifts or are partially unknown, it is important to be
able to verify the robustness of ... | i | 9d1fc544797dea02f66666ff9b71ad88 |
We have introduced an approach for learning conditional radiance fields from a collection of 3D objects.
Furthermore, we have shown how to perform intuitive editing operations using our learned disentangled representation.
One limitation of our method is the interactivity of shape editing. Currently, it takes over a m... | d | 84f637a3a29028b843683b0587c53fab |
For the pre-training task selection problem, a probably exciting direction would be to design pre-training tasks for a specific downstream task automatically, just as what Neural Architecture Search {{cite:555edf6794b7506180482603d1a66f333cc2cf7f}} does for neural network architecture.
| d | ef170385b23a787b2064afafa334a77d |
We compare BMCoGAN with recent state-of-the-art methods on generalized zero-shot learning, and the results are shown in Table REF . The results of LATEM {{cite:b65984cadc626af1d534c5ef0e62564aa5b092f5}}, DEM {{cite:a30d9af7f1966d4ffbf00617b9df241ba153ff95}}, and SGMAL {{cite:76f06917be1a8f647554a69f9b726c4166e283d5}} a... | r | 7e151a14123dcb16ccf87356e03cd294 |
The method consists of a few steps. Firstly, a fast enough process to
construct the data sets with which train, validate and test the NN is needed. In
this work we train the NN with a set of 2PCF mock measurements
obtained from log-normal catalogues. The construction of these input
data sets is described in detail in S... | m | b45bc413121b06e850d5f1fff265d621 |
Recently, several methods based on data-augmentation have been proposed and proven to perform well on a large spectrum of SSL tasks. The idea is to have a model resilient to strong data-augmentation of the input by predicting pseudo-labels for unlabelled data using a augmented version of the input {{cite:91cb3c6085626d... | m | 0a67e25ddec07bf8c61212c9a85cd9c9 |
Our preliminary experiments and analysis show that summarizing legal contracts in plain English is challenging, and point to the potential usefulness of a simplification or style transfer system in the summarization pipeline. Yet this is challenging. First, there may be a substantial domain gap between legal documents ... | d | 9f7a92abfb547d6b1b4a814027d97285 |
(3)
Branching ratio for the {{formula:35bc3a06-a94c-43ef-9bf5-10f60c546ff8}} {{formula:c4ab40ac-a28d-4b5d-b4ff-e340e1d531aa}} {{formula:bdf136b5-309e-4ed2-972d-7498acad4f0c}}
decay is a few times of {{formula:ea753af4-32da-4c7a-ac90-e6559647d1ad}} . So the nonleptonic
{{formula:abe4131a-6be6-49d0-981e-a2245b357734}}... | r | 8c8eaaf70c366dba1f40c6cd4e841017 |
Flickr30K Entities.
Table REF also reports the performance of our TransVG on the Flickr30K Entities test set. On this dataset, our TransVG achieves 79.10% top-1 accuracy with a ResNet-101 backbone network, surpassing the recently proposed Similarity Net {{cite:b508ab11f80790ccbcf76e6bf77a5c5eee2f37fe}}, CITE {{cite:a3... | m | 34d867de84aedea1d3e5b4c304dda901 |
For every {{formula:2254d87c-9219-4821-bf66-3813c6cc524d}} , the {{formula:ee967f6a-3e43-45ac-9c69-d408ac15bfcc}} -Weisfeiler-Leman algorithm is a generalization of the naïve colour refinement, giving an approximation of the {{formula:3c1621b7-73ef-41f3-ac8c-8caf0ad3e03f}} -orbit partition. For a graph with vertex set ... | i | da0e323576d812f842305972e6d26c9b |
As depicted in Figure REF , a more complete comparison between DeepHMap and DeepHMap++ is given. For all eight sequences from the Occluded LineMOD dataset {{cite:ff89a59e1a5ff1855d37f1f33af77feea69f4ced}}, DeepHMap++ steadily achieves better results than DeepHMap under different pixel thresholds. With the 2D reprojecti... | r | 3b9b32acf608fc46535d49488494bd29 |
We use AI Fairness 360, a package for the discovery and mitigation of bias in machine learning models. The protected attribute in our dataset is gender, while the favourable class is “man”.
We employ two classification algorithms implemented in ScikitLearn {{cite:a565572abf3eb9fa6c54de3f6c95c97b586899d9}}: logistic reg... | m | 59124113cdf1355682d00eeee6fbe6a8 |
The MW-Net Learning algorithm can then be summarized in Algorithm REF , and Fig. REF illustrates its main implementation process (steps 5-7).
All computations of gradients can be efficiently implemented by automatic differentiation techniques and generalized to any deep learning architectures of classifier network. Th... | m | 7ca718da0fe32863791714199cb748bf |
Perhaps closest to our work is Bacry et al.'s {{cite:e01c897073f5df3077f152203b54685a0d3f62f5}} study of mean field inference of Hawkes process values. In particular, those authors inspected the effect of varying the decay parameter across a range of values: With increasing decay, fitted self- and cross-excitations dec... | d | ff412ec4b60f538a3952c24b36d2eda0 |
Efforts in making legacy models couplable, either for MOSSCO or
similar frameworks, however, leads to additional benefits besides the
immediate applicability in an integrated context. Couplability strictly demands
for communication of sufficient metadata, which stimulates the quality and
quantity of documentation and o... | d | 305783a839e8e1767825adcf419bf5f5 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.