text stringlengths 54 548k | label stringclasses 4
values | id_ stringlengths 32 32 |
|---|---|---|
The use of active learning methods with deep learning frameworks has been investigated for applications such as image classification {{cite:f4384d564a298c428df0e1f3b20adf616c9f0ff8}}, {{cite:abe6dcd1576b0d99097b584b5c21fc273f6f0474}} tasks and natural language processing (NLP) tasks such as named entity recognition {{c... | m | 6b901f22d7755341339988f54c500f60 |
Approaches to quantum signatures of chaos fall into two categories. One involves investigating quantum variables that distinguish between quantum systems whose classical counterparts are integrable and nonintegrable.
These approaches typically look at energy spectra properties {{cite:763892cb62fdb85808393d5c47e7d1fb2d9... | i | cf073292ee64a147c010202fb4ce75c3 |
where {{formula:86c69b25-a64b-4329-bf29-e5b944f65e5c}} is the inverse temperature and {{formula:bec722d8-62e5-4a69-a792-49b55f4597b7}} carries the information about the primary operator perturbation of conformal dimension {{formula:9b882976-f087-4ded-8526-86b2224e74ed}} . The parameter {{formula:e96083ae-d9db-4886-b9... | r | c291e62e2ccf70c54fb6322f33c36dbf |
A close look at the temperature expression of Lifshitz-dilaton
black holes (see {{cite:6241fbeebb760c348fcc2ba34c7b2983b7f915e1}} and Eq. (REF ) of the present
paper), shows that it is nearly impossible to solve this equation
for {{formula:927cc957-cb7d-48fe-842a-6eb8e1d815d5}} (or more precisely for {{formula:8be4aff... | i | 6f95a7a245eceb7c962e32d9e82bc66c |
Proof of Part (a): As {{formula:308cefaa-b4c3-4538-82cd-ec5bb565f4f3}} for any {{formula:935fa3ae-8c10-49fd-bb59-fb95fb7b8ef0}} by Assumption REF , by applying
Theorem 2.1 in {{cite:1505a7628db1b9709f3923a69ebdd89752b1d85b}}
with {{formula:15b3c43d-7187-4f18-83f1-c8691f392f1d}} and {{formula:7e0a2935-cb0b-4bf8-80c0-... | r | 8d36de929540b1a6969d25f43557f2c0 |
A major drawback to the edge-on perspective is the more complex
geometry, which implies that each position {{formula:78b9eb08-9259-4530-9197-60c1a1aba401}} along the major
axis is a superposition of light emitted at a range of radii {{formula:074387f6-e27c-4d95-97a4-b2ceb5a9c38c}} .
Various authors have therefore set ... | i | e9cd28f61b367a1ef9723e9a802f862a |
NPs sample two sets of observations, for the context {{formula:995c2a46-60f7-4cb9-9e00-395d96031f3f}} and target {{formula:3914e647-ec05-4b32-a125-bd12cf06fab8}} sets. The context set is used to generate the representations and the target set to verify the predictions of the decoder. Our objective is to learn the und... | m | b8acaf0a5af7aa92c0f37e54944f7248 |
Recently, convolutional neural networks (CNN) have achieved significant success in many computer vision tasks {{cite:74751fbc37c0bb2333445adc14fd9e2e28e6a4c9}}, {{cite:3010440e543e2d883f1e5c870e241e7a4d415eb6}}, {{cite:0fc0a98c1325d01ba325a941289d2c68ebdab6f7}}, {{cite:1d03b264abdc3fae40a26841c3f247bd6990e8b8}}. Moreov... | i | 18527d32d49c3e457aabd0efb6978fc6 |
From Fig. REF we see that for
a {{formula:8f0ee860-7df2-49c9-a1e5-e843cda96d6f}} SN at a distance of 1 kpc, a 1-ton detector will yield, on
the average {{formula:8999e6ef-2e3b-46d7-89de-66d68eac8b0a}} 10 (20), 7 (6) and 6 (2) recoil events above 15, 20
and 25 keV, respectively, from {{formula:1058036b-fe44-466c-b2bb... | d | d987db77b54379c2e63076400069c8c4 |
Here, {{formula:7d5b5810-0bbe-4864-bf27-5c5a81ada6cb}} , for which we may utilize {{cite:6d70e55a61d4af874c6bb4b3dfef4fe3917d98de}}
{{formula:7c53db14-a712-42f3-8e81-c70cad2fbf8d}}
| d | 4d6c91d19bef093606eba9634e18536d |
Semiparametric model of ARPES intensity.
Now that the basic scheme of the Bayesian analysis is demonstrated for a single spectrum, we apply this analysis to the actual two-dimensional (2D) ARPES intensity through a semiparametric modelling of quasiparticle bands. For a testbed system, we have chosen {{formula:68306fa1-... | r | 48efb8ac62aadc546482319734f8e68f |
It is common to use actor-critic algorithms such as deep deterministic policy gradient (DDPG) {{cite:9cc297d9703b6a657db0dfc1acce3f9a0381a9e9}}, twin-delayed DDPG (TD3) {{cite:73f7bf68d113306e994d9bbfd018629257587f72}} and soft actor-critic (SAC) {{cite:6ad33aaee899e20cfa095d69d9629b22b836ac09}}, to solve multi-goal co... | m | d161631e03f1f487b3f3eae27ee696e7 |
The distribution of the sum {{formula:fc951f49-eaf9-4586-922d-e9972883793a}} ,
where {{formula:53fc61b4-c2cd-4434-96a9-12689660989c}} for {{formula:3ea33162-8b9d-4290-93c3-03f0db47d45d}} are not all identical, is called (general) hypoexponential distribution (see {{cite:5e7e8341d1b0527317b8e80c1b1ae15e8bf92ab1}}, {{c... | i | 91930562a79dec4f49af31fe7b7bb668 |
Another recently rediscovered exploration idea is Thompson sampling (TS) {{cite:24a7f3b7fc6a4b1183a060cc849dd6f0aba9eabd}}, {{cite:1695efbc4b941defa9952e64fcaa5872bb9a8e09}}. It is motivated by the Bayesian perspective on RL, in which we have a prior distribution over the model or the value function; then we draw a sam... | i | 3fa24e763a64b69a8b7267656c9a9235 |
We now propose an experimental evaluation of DP-SGLD and compare it to DP-SGD on a classification task using logistic regression on two vision datasets, CIFAR10 and Pneumonia, a dataset of chest X-ray images of pediatric pneumonia published by {{cite:aa3428fd4dd207b38160a0bef89b504e435cafcd}}. Details about the dataset... | d | 36ca1067459d34138d70cf1d1eed7dd8 |
We will first introduce a couple of known results for vorticity equations, for further discussions on the topic see {{cite:6d84d778ff7a01bc377cfc6edb9e70ae9acd584e}} or {{cite:3f905592d32ab283ede2bc44286dc620d4a0d6fc}}.
The link between the vorticity and the velocity vector field in equation (REF ) is uniquely establis... | m | 683c85ec4455761cabd4b5b201c3b358 |
Finite Volume Method {{cite:ec44ef2650a57204508085e12337bf10d8e17915}} adopts CNNs to generate coefficients for approximating spatial derivatives, followed by the standard finite volume method. CNNs are optimized end-to-end to best satisfy the equations on low resolution grids, and produce accurate numerical results: i... | m | 53848039aa020f150fd36aef1d8d0940 |
Since the papers above and our deal with higher spin interactions in {{formula:40b25b2a-9faa-494d-be80-f3d5de88f0a1}} and we used the notation close to those of {{cite:5c8adaf6e2cc3d5f4a22472594d1d94a5dc88ecd}}, {{cite:6730c65580e990e9bf0468d1f22da3fbcf658a74}}, {{cite:01186b096ba9946be2857a13d2a30f0e5382c2e0}}, {{cit... | d | b4e1c9a8b59107d77a389ae6ca3ad0f9 |
In our experiment, these changes in the ways participants integrate information when receiving prompts about unobservables did not translate into an overall improvement in human predictive performance. When interpreting these results, it is important to consider several characteristics of our study. First, the number o... | d | e90bd887441097dc9753cd07f689ec4b |
Finally, there is no reason not to apply our method to sets of phrases or paragraphs, by replacing CLIP with a summarization engine, such as those based on transformers {{cite:def00e720759e123719ee8eb60aa60310de728e9}}. Distancing ourselves even further from the current work, we note that with the advent of powerful im... | d | 6adc0c84ba335b5c0de92d39be171983 |
Although the prostate gland is represented in the image FOV center, since the organ to be imaged is always positioned approximately near the isocenter of the principal magnetic field to minimize MRI distortions {{cite:e5cdbb0abceb65b29a4d46fb0242a405c52f2184}}, sometimes it may appear shifted from the center or the pat... | m | 9a9f1ef24a0be73bd0f0e48756f11a6c |
{{cite:2a7468024d38888bc098d7b1295c0e8bbe5bfc46}} also analyzed observations taken of a field positioned along the plane of the Galaxy. These observations exhibited a hot component somewhat similar to what we see for our CGM fields, but {{cite:2a7468024d38888bc098d7b1295c0e8bbe5bfc46}} interprets this high temperature ... | d | a9b367ae11c7a954d43322f8183a507b |
Understanding fixation processes and their probabilities is essential to the analysis of evolutionary games on finite populations. In this context, {{cite:4c3d245502ba7d6d1edce8c4fb237b65e2cda687}} observed that even if individuals are interacting via the same game, the size {{formula:64fecc18-3e27-41e3-8e6d-548b25611b... | d | ead0e1d9c151b4c8112fedc6e00a464e |
Differentially private federated learning (DP-FL) has received increasing attention to mitigate the privacy risk in federated learning {{cite:a22112eb4b4173cbf38a6792488424e1f8dce3c6}}, {{cite:ff8fd2517ed333ee8add4ec28435f888ff798eea}}, {{cite:25a85584a113b3eed8294b244e69fc8bd78a7061}}, {{cite:3faff14f289dcedf9b25d59d1... | i | 1612656bb013fa46672a3ef07ac95dbf |
While decentralized learning is typically studied with heterogeneous datasets across workers, sparse (decentralized) averaging between is also useful when worker's data is identically distributed (i.i.d.) {{cite:fa3281097d9345cd88eed5dea4432f9664dca7a7}}.
As an example, sparse averaging is used in data centers to mitig... | i | b95e728e0c9b7429e8718861ab31d995 |
Case sensitive BLEU scores are reported in Table (REF ). Our model brings in an improvement of 2.3 BLEU over the LSTM model. When tested on lower-cased text (without altering the training procedure), we obtained a higher score of 26.2 BLEU. One major limiting factor of this analysis is that we use a small, fixed vocabu... | r | bcbb7a90ee787b246629e542fecd7591 |
It is interesting to think about the scenario studied here to the arguments put forth in
{{cite:5d4682998aa44cfca3219ce91b26f04139d88fdf}}. In that work, it is conjectured that a CFT with a large-{{formula:8d8f96d2-8c51-4d82-9648-a645c3b7be17}} expansion and a
sufficiently large gap has a local bulk dual description. ... | d | a99dccae0a85ea2743022d8a18309e80 |
In principle, the matrix elements {{formula:7450d68b-1174-4dbe-a69a-53a593f38026}} contain full information of the cross-correlations between stocks,
but carry also strong noises.
To sketch the backbone of the interactions between stocks, the edges should be greatly discarded with
respect to the complete graph. To thi... | m | 21278f19d721d14c8d2b78b7cbdb1b3f |
In particular, here we are interested in leveraging the EOB approach {{cite:1641edc6f1555183c749a42fe2009d2c7f499989}}, {{cite:c50b33e460bfa6c6d8a6f651b991bbd52ec263e9}}, {{cite:e5455ae04e76f1bd8a0ef36095d50e4cc6ab82f2}}, {{cite:b1243a128b97848d91451c3f0c2ebf0998cb50dc}}, {{cite:5f76b9f3294d307de453d219d51e88b7e5a81e48... | i | ea1aad3a8d36a128f4efb5a229373c32 |
Finally, we show qualitative results across different benchmarks, namely the noisy point cloud variant of SHREC'19 mentioned in our main paper, additional qualitative examples of scanned point clouds from CMU Panoptic dataset {{cite:199f96fcc6f4ac3d8341a8de0b8647994db2b8df}}, animal shapes from Deforming Things 4D {{ci... | r | f92aa60ec607b6e9e92f1d53da2dbf0e |
Consequently, the estimate of {{formula:69cea750-d3ff-44bf-b7d2-5a9d8dec046c}} requires the source term to be bounded, which is not the case in general because the velocity field is not Lipschitz in the framework of Yudovich solutions. This suggests that all the scenarios of well/ill-posedness are likely possible for ... | i | 2479c9700494a841417972f84c0c8fa2 |
Lemma 2.4 (Lemma 2 of Section 3 of Chapter VI in {{cite:3993601a78740a917c3356f9a709a4d1513a4b75}})
Suppose {{formula:3a2cf547-49fc-485e-b158-8858044d2656}} , where {{formula:da8ed8be-3cf2-4a70-9b7f-2740f7be07dd}} is a Lipschitz function satisfying: {{formula:dc61a554-870c-492a-9020-20fae575c317}} , for all {{formula:... | r | 8d8f22fe2a97e6c39d73594e7cd403c7 |
In this work, we propose a NeSD framework to address the sound field decomposition problem. The NeSD approach is inspired by the neural radiance fields (NeRF) {{cite:3d505af889ca04f684822b1c75dfd97d848c75da}} for view synthesis in computer vision. NeSD has the advantage of predicting the locations of wideband, non-stat... | i | d93737791e1d0984d6a5f0cf8fa2670a |
Mobile edge computing (MEC) provides a promising solution to enabling cloud computing services in mobile edge computing-based Internet of Thing (EdgeIoT) {{cite:acd38b004458438e281da161df29c6b0f12761be}}, {{cite:274e8719becddb0cca9d372cb94d9c766352c850}}. The IoT devices can offload their local computation-intensive ta... | i | 9197ebbc9beaa3c788e459ea74eb4261 |
Quantization of the weight and activation of the deep model has been a promising approach to reduce the model complexity, along with other techniques such as pruning {{cite:328faa6f0adf950400f9bc87f2d2c6521d9c9cd2}} and distillation {{cite:d10c2f4802d34c5dfaaa39286af384e9b55c45cb}}. Previous studies, both on weight-onl... | i | 07b79767f96d0c82b57551c915801eb3 |
Datasets. We curate natural backdoor datasets from two popular open-source object recognition datasets: ImageNet (released under a BSD 3-Clause license) {{cite:fd402fb4bb8094876821937376bbcd9d489a2d08}} and Open Images (released under an Apache License) {{cite:bf67c90670c433c84620c6a22579a22fe4d2d7b2}}.Note that approx... | m | 38141982615dd5759ca4dd73d4252d1b |
Verify the Certified Robustness: We run experiments on synthetic networks to verify the certified robustness; the synthetic networks include Barabási-Albert (BA) {{cite:3abb1967b972b571aa503e4e88814665ce55bb6e}}, Watts-Strogatz {{cite:e28edbbd10d60d072f1b0cb595ef6460f4ed04ea}}, and Block Two-level Erdős-Rényi (BTER) ne... | d | edb06d3dae501202c6c5de6f72b68f1c |
(2) The content preserved style mixing aims to add style noise of the other domains to one image while preserving the content information. While there exist some popular mixing methods (e.g, mixup {{cite:ea77f72fa3ee1d21ada51062695df85768348c91}} and cutmix {{cite:0ec428abef24efa818ba91602246daa73ba12230}}) able to mix... | i | a8aefb77471aa4cea0a5fa85cdb36628 |
The total corrections of exciton-LO and SO phonons coupling on exciton binding energies for the monolayer MoS{{formula:d3b4ed38-517f-4651-ad17-ee52bf3c9cce}} /SiO{{formula:b992e906-f4f4-4fc0-a89e-8c533358e041}} system are presented in Fig. 4. It can be seen that the magnitude of exciton binding energy can be tuned in ... | r | 00088dc5c26a8563416607c3e6cfb84b |
Most of the work done on sparse linear regression solves (REF ) heuristically by replacing the combinatorial condition {{formula:76974997-71f5-47b6-b52d-13b36e6cc3e7}} by a {{formula:1bd05cd0-684b-44bc-b891-0a4621e7caca}} -norm constraint {{cite:65b4a70e0d3c2c5cb2ccce5a651fa05ed23766c6}}.
Elastic Net or Lasso is usual... | i | 1bd5ffae7bd6320188e783486041e6be |
This study focuses on pruning for the increasing need to accelerate the application of DNNs. Pruning accelerates DNNs by deleting the trivial elements in the weight matrix. The importance of each element is evaluated by the importance score {{cite:b130dd7ddef94503c487e5840a0b911a9b2baaa4}} so that the elements with sma... | i | d0f3344b062b6776fff2c2b22bd6fccf |
Direct training methods were proposed to address the aforementioned issues in biologically plausible training algorithms and DNN-to-SNN conversion.
These approaches, which are based on supervised learning, use surrogate gradients {{cite:c9512eab7b159361029306dbe70bdef853440027}}, {{cite:28256ceeabad4e363c32a1378fc1be8c... | m | 77210822452c2af4bc456d6ad91ada5c |
Naturally, this paper is also related to the rational inattention literature pioneered by {{cite:32247a9ff4ad041819663aca5c873cea9f8848b1}} and furthered by, e.g., {{cite:48862b45b21d0b1cc74ec504a838cb61f97c3cac}}, {{cite:6684ecb70e99c34d60aac455899d4cb7a02802fc}}, {{cite:5c6c99dc715a1c5af54cdb826a82bb022c41aca5}}, and... | d | 4ad2fa4f6d7c01ec77f37db353031758 |
There are several mechanisms that could drive the eccentricity evolution of a SMBHB. For instance, at sub-parsec scales a binary formed by a galactic merger may be embedded in a dense stellar environment. As discussed in {{cite:db0e35f8028328bcd4245e7e1892c65a154bee9e}}, if one assumes an isotropic stellar distribution... | i | c9bb01b6e463c9df6f46966086a1ae77 |
The failure of this model offers insight into the ability of new physics to resolve the Hubble tension. In this case, the pressure fluctuation drives the growth of the heat flux on subhorizon scales as shown in Eq. REF . This sets off a cascade of effects, softening the gravitational potentials, shifting the acoustic p... | d | 30bb1849712be050a9c271380deb44b2 |
Table REF presents the performance of our approach across all phases along with several ablations. We find that our approach is able to consistently solve the tasks, is robust to observation noise, and can recover from failures (i.e., dropping the object). However, there is still room for improvement. We often find th... | d | 9f435e9fff2c9232de850a41f4f5d6a4 |
where the eigenvalues of the Jacobian of the vector field {{formula:95156ea6-1638-40ea-818e-1f5c87b97dbb}} are located in a narrow strip around the negative real axis. In order to study the stability of Runge-Kutta methods, it is common to consider the linear equation (see {{cite:1c4391f55f46648ccb85f3ceec84bf8cd0eb08... | m | 7bcde859427d252545483dd1178d001c |
Most hypothesis-driven neuroimaging analyses depend on specified models when proposing a statistic and fitting parameters of the GLM. This is a major advantage when data (nature’s mechanism) is drawn assuming a Gaussian distribution (model’s mechanism), and the inference drawn from the experiment may be misleading {{ci... | d | f116b48fe732ac193777e4c3efec44c9 |
Considering the variety of coupled-spin-chain compounds, including the other materials in the {{formula:9fe484c0-ff06-468f-ab00-a1810e022d6d}} -type hexagonal perovskite family {{cite:0daaec1d6d69875cedbe5aa2a9ecc1b0a70b79e3}}, {{cite:ec3de9b6fce655594a131f618ef52f9b55de7eb2}}, {{cite:089fa9f58a9456ca751c7b94f91e78c497... | d | ae4e6eceb655d267c62b85c4a303b1cd |
Within the model-based Bayesian framework for causal inference {{cite:748e1939e9f812cddfd61781309003a70f8a1686}}, {{cite:b4ceb4f8c67e3078613fdbd3ee41f7e6839ceebd}}, the {{formula:5034e08e-c3d0-4651-80d8-0fbdf46806bc}} are considered unknown parameters. The goal is to sample from their posterior predictive distribution... | m | faaa96f62b54e9b1974fe176cc03699b |
Inspired by Domain Adaptive Faster R-CNN in {{cite:2b0d62471471eefa1004c545fcb993cb413f0356}}, {{cite:b390166a55629fae76c921d3f60b80da09cf2e48}} adopts a similar idea and designs a Domain-Adaptive Mask-RCNN (DMask-RCNN). The source domain takes the clean images in the MS COCO dataset, and the target domain takes unanno... | m | 3e923058a483fe446eab151cdefc8089 |
In this work, we propose to generate 3D portraits that can be explicitly controlled by allowing the user edits upon a group of semantic parameters. To this end, we try to make the best of both the explicit parametric model and the neural face representation: the 3D Morphable Model (3DMM) {{cite:272fd2ef3e40b411e1395135... | i | a518c9a48c0d23051593567866becef7 |
In a cooperative interaction, the environmental noise is usually expected to lead to a detrimental impact on the emergence and stability of cooperation {{cite:3a0390a583208aa121c4329cccfc0f0ab79a2179}}, {{cite:55666dcb902b844c45cf9850139e203a16a1bdc6}} and thus requires additional supporting mechanisms such as apology ... | d | 08edfde3c7d7775b5f198698a6473db0 |
Combining optimism with MD and FTRL yields their
optimistic counterparts, optimistic mirror descent (OMD),
and optimistic follow the regularized leader (OFTRL), respectively.
OMD and OFTRL both provide tangible benefits when problems are not quite adversarial. For example,
faster convergence to a saddle point on the av... | i | 4e236c9a2fef100ce0e330abed19f4aa |
Carefully measuring accuracy allows us to disprove the claim {{cite:3da2c045623a836477878b46ca0f22c696162837}} that sampling simulation parameters from the posterior, as is targeted in a wide range of algorithms {{cite:285434220ac4b86978451cb28a07fd52734a5c91}}, {{cite:4e6271709a9690967a1f213a4eb8ef8e35b1deb0}}, {{cite... | d | 99d5287111e505576dd9403712336498 |
Beginning in the early 2000's, a few authors addressing general methods for machine learning and artificial intelligence included the CVM in their comprehensive reviews of methods. For example, Pelizzola broadened his earlier (1994) {{cite:55d6231e39c6c91c61051a0705bc53330e0fd4c7}} work to include the CVM in a more gen... | m | ce3c49854ee7fc2144bc798ca4f0c8b2 |
We consider that our theory can be experimentally verified.
In particular, the dynamics of isolated non-integrable many-body systems have been investigated using artificial quantum systems such as cold atoms {{cite:4f422907e7033f6e71e46ea311bfbbe2160a47d7}}, {{cite:af3238f50cdead58b0358871c6bab8c09fc5a407}}, {{cite:4e2... | d | 1e848cbc557b7d4fdc188d5d1197f620 |
SPARF View-Generalization benchmark.
In Table REF , we report the average PSNR results on the validation set of SPARF for different methods on unseen shapes during training on all 13 different object classes and on out-of-distribution views. It shows that our SuRFNet can generalize to out-of-distribution views on unsee... | r | c7ba77c3080e56dde34ec72df180505a |
In this paper, we introduce a quantile estimator which is computed by directly pooling the detailed simulation outputs from various replications to obtain a sample quantile estimator,
called pooled sample quantile estimator. By pooling the dependent (within each replication) and independent (cross different replication... | i | f374077090ae7b8726b5ed39a70c778a |
In the following, we will give an example of the inconsistencies and inaccuracies introduced by the time-correlated noise using the case of TTVs. TTVs may be caused by a perturbing object in the system such as a stellar mass companion of the host or another planet{{cite:74723366c66c994c1df338fbf4715bd8a14bcbaf}}, {{cit... | i | ce0e5486f84748b8361c8a1a450c5d4a |
In this section, we describe our approach to learning fair and accurate models and highlight the connection to MTL {{cite:43f0cddaa1410bd7aa6a02376a8c90d550d19b5f}}.
We consider the following functional form
f(x,s) =
w(x,s), (x,s) X S,
where {{formula:b2da4d80-2ff3-487e-82de-8da5e54d4855}} is the inner product... | m | 81abadcd6b7a3d244fd783b6a0760c1f |
Model Uncertainty:
The Dice Loss {{cite:0a8f4731ff40e2cb33cad6bdaf700c23037afa4f}} between prediction logits {{formula:a824c5d2-ce9e-4e48-8f90-adc57c83b25a}} of the DOINN model {{formula:792d22f9-2a53-491f-bfc3-0a08381465fd}} captures the model uncertainty:
{{formula:39ecfacd-ae3b-4486-aad0-46d7d7f24962}}
| m | b507c272f6d3449dca769e92641e784c |
Sum-Product Networks (SPNs) {{cite:b0133db938d84ad1e9ecf9298ad5c28e644d0b54}} are deep probabilistic models that permit exact and efficient inference.
SPNs model probability distributions as a computation graph that consists of sum nodes (representing mixtures) and product nodes (representing independent factors). Spec... | i | a7f00e1e730136e2a1c306e27b995574 |
C. Binggeli would like to thank A. Gavel, A. Lavail and M. Sahlén for helpful discussions. M. C. Toribio would like to thank A. M. S. Richards, E. Fomalont and L. T. Maud for useful discussions. A. K. Inoue and K. Mawatari acknowledge the support from the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant ... | d | f76cd6041d51d6c56890733e462407bc |
On the other hand, we have not tackled the most important questions: do asymptotic observables exist in expanding spacetimes? If so, what are they, and how are they computed? Can census takers in an eternally inflating universe be used to define a measure on the string Landscape, and if so, what is it? It is very possi... | d | 8057a49387077fc1ad9c119609790f36 |
Several methods on properly constraining a weight matrix of a simple RNN without gating mechanism are proposed to alleviate the gradient vanishing problem {{cite:672cdb0b15ab9079f5eae3a796bf1ef24844b66f}}, {{cite:cf28c4331d803b4ecd7b9632e16c83c92764908d}}, {{cite:8c615af154a10c7c5cd062dfc425adc9de392b36}}, {{cite:9c106... | d | a7352d3a9e2c5f4dbbc70f9bf35b02e9 |
A cognitive theory is a general postulation of mechanisms and processes that are globally applicable to families of tasks and types of activities rather than being dependent on a particular task. Cognitive models are very specific representations of part or of all aspects of a cognitive theory that apply to a particula... | i | f03830e4fb69289268706f37a138ce36 |
[leftmargin=*,label={{formula:88588987-a56c-49ac-9c62-7928db1853a6}} ,topsep=2pt]
ViP-CNN {{cite:9062b03919aae372fa054b1833f690405517ce54}} predicts predicates using visual features from three bounding boxes, including the two object bounding boxes and a tight bounding boxes covering the union of the two objects.
PPR-... | m | fb0ed2fd23a2250c3f8b594f12ddb834 |
The experiments of Relativistic Heavy Ion Collisions (RHIC) have
produced a strongly-coupled quark{{formula:6f8c8a6f-1454-4376-a1b6-d25ef71da859}} gluon plasma
(QGP){{cite:3478c6f0245df8678c33938912225f2953775fe9}}. There are no known quantitative methods
to study strong coupling phenomena in QCD which are not visible ... | i | a00fb5f787a204f514b8066116c9f270 |
Diversity in a small world.
A network has the “small-world” property, if it is possible to go from one vertex to other vertices by passing through a very small number of intermediate vertices. More precisely, the small-world property refers to networks in which the average distance between vertices scales logarithmical... | r | 1c35f4c7fb1dcf1518140db450866d33 |
We report simulation
results in R with synthetic data demonstrating excellent agreement between our theoretical
predictions (Theorems REF -REF ) and the behavior in practice. The code can be accessed in https://github.com/philipthomp/Outlier-robust-regression. For robust sparse linear regression and trace-regression pr... | r | dabe85263b6ba0b610a3c3502c81eb4f |
In LQG, the dynamical equations are often too involved to be solved explicitly, making the identification of physically relevant subsystems and the calculations therein so far impossible.
It is therefore judicious to test many concepts and techniques of LQG in symmetry reduced systems, where this issue can be overcome.... | i | cdbe3f2951c1b13dcdb953d7e294a196 |
Formulating the environment as a player {{cite:3277c7430062c0bca8f30339946c7fc0216ceb68}} in an agent vs environment game is an interesting way of ensuring the distribution of tasks (strategies available to the environment player) does not bias the ratings of the agents training on those tasks. In Sections REF and REF... | d | dcd67a80c652d9f30a672a15a117db4c |
Then the equation has a groundstate solution {{formula:8609bfe5-4e75-4285-8bf0-b1d082a0682e}} if {{formula:99b40d3e-0194-42d5-a3cf-0fc54aaae624}} satisfies strictly the inequality in (REF ) (see {{cite:643cc33ade5f36eecfe31801998e96aa336c0c63}}, {{cite:5c9be038a66fa4b9fb472364aff0b82ed71db4a2}}, {{cite:d943f0e9801e59... | i | 31e430cc99ed27982e32fb40b9dd58b2 |
The dCME can be approximated using the Fokker-Planck equation (FPE)
and the chemical Langevin equation (CLE). These approximations are
not applicable when copy numbers are small {{cite:502fbbd0e80487bb404ecfe92ddd41b77b3d2828}},
as relatively large copy numbers of molecules are required for
accurate
approximation {{cit... | i | cdf1cf7c7a09050d65298910914b3dab |
In this section, we report the results of three different types of features for the three training methods on the two datasets: (i) Maximum Response (MR) filter and texton maps of {{cite:b1f12233296825dd6baa656a68ea912c223ccdec}} as baseline handcrafted features. We used a publicly available implementation of this meth... | r | c8eea91206f4a70e527d687d0e9e6fcc |
The effectiveness of adjusting time-constant parameters to the task may also have implications for neuroscience: though effective time-constants of real spiking neurons are variable and dynamic {{cite:daf97a2dd92f83b326086ed2aea1b5e3194c935d}}, the benefit of training these parameters in SRNNs suggests these neural pro... | d | 2fdc11a28173aaa99c8dfe7911f0333b |
Multifractal analysis is a useful method to quantify properties of complex system, and it has been applied in many different fields, e.g.{{cite:3841f60980ef32c147c8403d1a50be109b3c1cec}}, {{cite:542a581c515874c4ddc2e175c247729a1062d168}}, {{cite:37f9c7046cfbd3d6ed025bc1452f391d22e9d056}}, {{cite:336469178236693b62ca41d... | m | a8adbb37af8fab204e79f798cd4491f4 |
We have upper bounds for Euclidean {{formula:1bcd67b8-590a-4583-b1e4-d49dc066ca0d}} -distance sets with several conditions, which are counterparts of that of {{formula:327a1259-f580-43d2-9499-472207461319}} -inetersecting families. Let {{formula:0d47fa7b-51cf-4d59-a760-ecdac2e52b00}} be an {{formula:6f524cbd-e83c-4dcf... | i | 86bd533d117aedbf5033d19a067a3ebf |
Utilizing the feature extraction capability of deep neural networks, Simonyan et al., {{cite:d9830ed514c8f8ce0dc7cac0604d316108feaafc}}, uses a 2D CNN to create a holistic representation of each input frame of the video and then uses those representations for recognition. Temporal dynamics of a video can be modelled by... | m | d968614c9a0960be9c529497e5b55f3b |
For the first statement, use that {{formula:ac45c697-8b0c-4283-934c-cd79ea9f50d9}} to write
{{formula:275c2034-6607-41ae-ad4d-e589567289dd}}
for all {{formula:9361423a-eca3-4cf2-8895-9007c33749f7}} , {{formula:3b148264-7aef-4938-9ea3-5b31c0b2e18a}} and {{formula:6306116b-ccee-429d-9651-bb23a4d2d7dd}} , where the fi... | r | 0be793f1b35dbb31b3ed8679c1ebb03f |
Empirical studies. Coincidentally, shortly after a preliminary version of the present paper was submitted, NeurIPS 2021 required all authors to “rank their papers in terms of their own perception of the papers' scientific contributions to the NeurIPS community.” Using this dataset, an empirical study may be used to ana... | d | 641c63e30093570c9b5c4859646a40f4 |
In this study, we aim to design a paradigm that distinguishes the causation edges from spuriously-correlated edges and obtains the faithful DAG.
Although learning causal structure from observational data is challenging, we draw inspiration from invariant learning {{cite:c7f91af3b68c6f4479821d94c5b72b19942e12e9}}, {{cit... | i | fd0d542b1a2f532e1f490860a86a4a53 |
Here we adopt an algorithmic information theory {{cite:62168e958ed257d79d44e281adee08f73aa810c5}}, {{cite:d4ddeed1f7864009565062cf220ea22c180a90b0}}, {{cite:b26772f4ce5cb8078f87dfff58fa46d361e3b95d}} approach, and argue that if the objective function to be optimised is `simple' (in a technical sense, discusses below) a... | i | 3b254cd87ac64f037163ed8e94a0d47b |
Answer these interesting and vital questions about particles is the starting motivation of the current work. In the Standard Model (SM), we describe the excitations of relativistic quantum fields that are characterized by fixed quantum numbers including mass, spin, and various charges {{cite:a2eef95e79be04222b8aa402148... | i | 018f0eb7c8df28fe32f696a7dcd88786 |
where {{formula:4d332d0d-e0f2-478d-8555-7052bfe54562}} are transforms that aim to maximise the correlation of the vectors. {{formula:db1cab6b-a7e8-4078-a66b-6ebe6fda4931}} increases where neural representations have encoded more similar information. Further details regarding SVCCA analysis can be found in {{cite:08bb... | m | 4771aa6b1bf60bba7e82badd9c9ad64f |
The investigation of all possible adatom species in rows two through five of the periodic table was completed using JDFTx {{cite:9c6e33c12f1e589193898611f87ef232055c406e}} to take advantage of GPU functionality. These calculations were carried out with ONCV pseudopotentials {{cite:d86805e37a5237f0e6c4b94eb956a3765239a2... | m | 4280984e38b5a09ba91e37a2baf16e95 |
The first-principles calculations were carried out based on the density-functional theory (DFT) as implemented in the Vienna ab initio simulation package (VASP) {{cite:9561e43a1abd7caa9b28d147d81e09d1a7ac40a0}}, {{cite:76591b7139dbbf32bc9bf09594bbdaf45c322455}}, using the projector augmented wave method {{cite:cddb492c... | m | 66d1a21b978bac4865b2274ad4ef0840 |
With the very recent release of the third GW transient catalog {{cite:55ba8594c12b404a7138d4de616d17b411e8972f}}, the total number of reported coalescences increased to 90.
Some of the more remarkable events detected to date include:
| i | a95740b2fa6e5ecac5ff0a03fd67f123 |
We did not provide an effective analog of the global Clifford unitary transformation-based method in {{cite:4c87b4ea76068ca56d561ca73600c810cf56b122}}. There has been work which provides description of global alternatives using stabilizer states {{cite:ea10f3f9052959210098c94f3025969366df19f8}}. Whether there can be a ... | d | f62fc7318410548a9fcd4eef76503c52 |
NVM devices present an attractive option for implementing synaptic storage due to their demonstrated potential for low-power multilevel operations and high integration densities {{cite:d424466734a5ffe1cfdaa22c315ead62f26d7099}}, {{cite:4019e5b04097ed8074b26cc6c3bda0dce4482cb4}}, {{cite:6f6c9232ee56c652127b0d43706af1f29... | i | 41b068c7071cfcca8d7421f19b3cd80c |
In conclusion, we have shown using quite a general treatment that Single Field Inflationary Models can indeed be compatible with the Swampland De Sitter and Distance Conjectures in a wide class of cosmologies where the Scalar Field follows the usual Klein Gordon Form. We began by discussing in brief about the Swampland... | d | 5a8452177338f74f9fc836462eeeb4d3 |
In the following sections, we describe our modification to LaserNet {{cite:3b227734be7097b807ad1903f0a8a561a7cefb8c}} to fuse RGB image data, and to jointly perform semantic segmentation of the 3D point cloud in addition to 3D object detection.
An overview of our proposed method can be seen in Figure REF .
| m | 1509d21c2e91df4716b85efd0cb0aaf8 |
Since the discovery of the {{formula:643c2654-2c51-4dd1-b4d2-0d52d6ca9fd3}} state {{cite:caec8dd44fef58371e137465c77c817937203446}}, over thirty
exotic hadrons have been observed by several experiments (see
Refs. {{cite:d74b0eac473bef38260afd8e9d4c6b78745299d4}}, {{cite:6f4159110838e2beff82d556c2700169e58c9add}}, {{ci... | i | b94a2d9c0f1f0a973026937c84036efa |
where, in our case, the acceleration {{formula:6bc288db-5479-42e3-8864-d0e9c22cac6e}} on each particle is evaluated self-consistently by direct sum over all other particles, {{formula:10aa9d06-2606-45a9-af4e-e930208d01ee}} is the dynamical friction [{{cite:ae88db193d429c2eecfe02abea061208c07dd11b}}, {{cite:13a2159eae... | m | 44914e08e00337825b75ad4b37c84e72 |
As for the Coulomb gauge, the instantaneous potential has too large
linear part, which gives an upper bound on the static potential
{{cite:0bf7af14e083d9c175c8d8ca1f249847e3beb6d0}}. It has been suggested by Greensite et al. that
the energy of the overconfining state is lowered by inserting
dynamical gluons between (an... | d | 7d399357c64451a31f919504cf67796f |
It also seems plausible that factors beyond stochastic optimization drive human development. There is, for example, evidence suggesting that children gain access to additional computational resources during their development, allowing them to apply more complex strategies {{cite:78365a6dcb325e2a7c78cc1094fc78db46136f8b... | d | cc93d5ee03c2170d3e53523eff0b1049 |
Effectiveness of Disparity Map Quality.
To quantitatively compare the reconstruction results of Stereo2Voxel and Stereo2Points with different disparities, we replace DispNet-B with SGBM {{cite:ccb8c002b9f357314d5cc9f1c92800d3b00dead9}}, DispNet, and GA-Net.
As shown in Table REF , the reconstruction results with dispar... | d | 34cbf799904e6d6d972d5f0869e19b25 |
However, most of those previous work can only handle regular data, i.e., they need to assume either the underlying data distribution is bounded or sub-Gaussian, or the loss function is {{formula:a16f056c-106b-4d97-9164-a84448151340}} -Lipschitz for all the data. This is particularly true for those output perturbation b... | i | f19fe577b9656cd1d5d3c6a5b06df356 |
We provide a first overview of the method in Fig. REF . As for the MOEA, we use Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) {{cite:dbefdddef46395cc3ae4ee8ece4785112c57557f}}, which has shown good results on this problem in prior work {{cite:e5115fbd1ff44cf9b4b8a0f146467c4b317b27ba}}, {{cite:9dfcd87eca93cc78a37... | m | 3c879392ce8cba62950c32c1106a8c7d |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.