Search Results for author: Nima Dehmamy

Found 12 papers, 5 papers with code

Latent Space Symmetry Discovery

no code implementations29 Sep 2023 Jianke Yang, Nima Dehmamy, Robin Walters, Rose Yu

It learns a mapping from data to a latent space where the symmetries become linear and simultaneously discovers symmetries in the latent space.

Generative Adversarial Symmetry Discovery

1 code implementation1 Feb 2023 Jianke Yang, Robin Walters, Nima Dehmamy, Rose Yu

Despite the success of equivariant neural networks in scientific applications, they require knowing the symmetry group a priori.

Inductive Bias Trajectory Prediction

Symmetries, flat minima, and the conserved quantities of gradient flow

1 code implementation31 Oct 2022 Bo Zhao, Iordan Ganev, Robin Walters, Rose Yu, Nima Dehmamy

Empirical studies of the loss landscape of deep networks have revealed that many local minima are connected through low-loss valleys.

Faster Optimization on Sparse Graphs via Neural Reparametrization

no code implementations26 May 2022 Nima Dehmamy, Csaba Both, Jianzhi Long, Rose Yu

In mathematical optimization, second-order Newton's methods generally converge faster than first-order methods, but they require the inverse of the Hessian, hence are computationally expensive.

Symmetry Teleportation for Accelerated Optimization

1 code implementation21 May 2022 Bo Zhao, Nima Dehmamy, Robin Walters, Rose Yu

Experimentally, we show that teleportation improves the convergence speed of gradient descent and AdaGrad for several optimization problems including test functions, multi-layer regressions, and MNIST classification.

Second-order methods

Accelerating Optimization using Neural Reparametrization

no code implementations29 Sep 2021 Nima Dehmamy, Csaba Both, Jianzhi Long, Rose Yu

We tackle the problem of accelerating certain optimization problems related to steady states in ODE and energy minimization problems common in physics.

Automatic Symmetry Discovery with Lie Algebra Convolutional Network

1 code implementation NeurIPS 2021 Nima Dehmamy, Robin Walters, Yanchen Liu, Dashun Wang, Rose Yu

Existing equivariant neural networks require prior knowledge of the symmetry group and discretization for continuous groups.

Lie Algebra Convolutional Neural Networks with Automatic Symmetry Extraction

no code implementations1 Jan 2021 Nima Dehmamy, Yanchen Liu, Robin Walters, Rose Yu

We propose to learn the symmetries during the training of the group equivariant architectures.

3D Topology Transformation with Generative Adversarial Networks

no code implementations7 Jul 2020 Luca Stornaiuolo, Nima Dehmamy, Albert-László Barabási, Mauro Martino

Finally, we compare the results between our approach and a baseline algorithm that directly convert the 3D shapes, without using our GAN.

Finding Patient Zero: Learning Contagion Source with Graph Neural Networks

no code implementations21 Jun 2020 Chintan Shah, Nima Dehmamy, Nicola Perra, Matteo Chinazzi, Albert-László Barabási, Alessandro Vespignani, Rose Yu

% We observe that GNNs can identify P0 close to the theoretical bound on accuracy, without explicit input of dynamics or its parameters.

Separation of time scales and direct computation of weights in deep neural networks

no code implementations14 Mar 2017 Nima Dehmamy, Neda Rohani, Aggelos Katsaggelos

We then show that for each layer, the distribution of solutions found by SGD can be estimated using a class-based principal component analysis (PCA) of the layer's input.

Cannot find the paper you are looking for? You can Submit a new open access paper.