Zobrazeno 1 - 10
of 1 360
pro vyhledávání: '"Ravanbakhsh, A."'
Semantic segmentation is a core computer vision problem, but the high costs of data annotation have hindered its wide application. Weakly-Supervised Semantic Segmentation (WSSS) offers a cost-efficient workaround to extensive labeling in comparison t
Externí odkaz:
http://arxiv.org/abs/2405.06586
This paper provides a review on representation learning for videos. We classify recent spatiotemporal feature learning methods for sequential visual data and compare their pros and cons for general video analysis. Building effective features for vide
Externí odkaz:
http://arxiv.org/abs/2405.06574
Autor:
Akhound-Sadegh, Tara, Rector-Brooks, Jarrid, Bose, Avishek Joey, Mittal, Sarthak, Lemos, Pablo, Liu, Cheng-Hao, Sendera, Marcin, Ravanbakhsh, Siamak, Gidel, Gauthier, Bengio, Yoshua, Malkin, Nikolay, Tong, Alexander
Efficiently generating statistically independent samples from an unnormalized probability distribution, such as equilibrium samples of many-body systems, is a foundational problem in science. In this paper, we propose Iterated Denoising Energy Matchi
Externí odkaz:
http://arxiv.org/abs/2402.06121
Autor:
Trang, Thuan, Ngo, Nhat Khang, Levy, Daniel, Vo, Thieu N., Ravanbakhsh, Siamak, Hy, Truong Son
Triangular meshes are widely used to represent three-dimensional objects. As a result, many recent works have address the need for geometric deep learning on 3D mesh. However, we observe that the complexities in many of these architectures does not t
Externí odkaz:
http://arxiv.org/abs/2402.04821
Autor:
Kaba, Sékou-Oumar, Ravanbakhsh, Siamak
Using symmetry as an inductive bias in deep learning has been proven to be a principled approach for sample-efficient model design. However, the relationship between symmetry and the imperative for equivariance in neural networks is not always obviou
Externí odkaz:
http://arxiv.org/abs/2312.09016
Autor:
Akhound-Sadegh, Tara, Perreault-Levasseur, Laurence, Brandstetter, Johannes, Welling, Max, Ravanbakhsh, Siamak
Symmetries have been leveraged to improve the generalization of neural networks through different mechanisms from data augmentation to equivariant architectures. However, despite their potential, their integration into neural solvers for partial diff
Externí odkaz:
http://arxiv.org/abs/2311.04293
Weight-sharing is ubiquitous in deep learning. Motivated by this, we propose a "weight-sharing regularization" penalty on the weights $w \in \mathbb{R}^d$ of a neural network, defined as $\mathcal{R}(w) = \frac{1}{d - 1}\sum_{i > j}^d |w_i - w_j|$. W
Externí odkaz:
http://arxiv.org/abs/2311.03096
Autor:
Jain, Vineet, Ravanbakhsh, Siamak
We present a novel perspective on goal-conditioned reinforcement learning by framing it within the context of denoising diffusion models. Analogous to the diffusion process, where Gaussian noise is used to create random trajectories that walk away fr
Externí odkaz:
http://arxiv.org/abs/2310.02505
Autor:
Mondal, Arnab Kumar, Panigrahi, Siba Smarak, Kaba, Sékou-Oumar, Rajeswar, Sai, Ravanbakhsh, Siamak
Equivariant networks are specifically designed to ensure consistent behavior with respect to a set of input transformations, leading to higher sample efficiency and more accurate and robust predictions. However, redesigning each component of prevalen
Externí odkaz:
http://arxiv.org/abs/2310.01647
We present a natural extension to E(n)-equivariant graph neural networks that uses multiple equivariant vectors per node. We formulate the extension and show that it improves performance across different physical systems benchmark tasks, with minimal
Externí odkaz:
http://arxiv.org/abs/2309.03139