activatedgeek / tight-pac-bayes
Code for PAC-Bayes Compression Bounds So Tight That They Can Explain Generalization, NeurIPS 2022
☆14Updated last year
Related projects: ⓘ
- ☆13Updated last year
- Transformers with doubly stochastic attention☆40Updated 2 years ago
- Official code for "Accelerating Feedforward Computation via Parallel Nonlinear Equation Solving", ICML 2021☆25Updated 2 years ago
- A modern look at the relationship between sharpness and generalization [ICML 2023]☆42Updated last year
- ☆13Updated last year
- Code for "Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance", NeurIPS 2022.☆16Updated last year
- ☆10Updated 2 years ago
- Codebase for Mechanistic Mode Connectivity☆12Updated last year
- SGD with large step sizes learns sparse features [ICML 2023]☆31Updated last year
- Supporing code for the paper "Bayesian Model Selection, the Marginal Likelihood, and Generalization".☆34Updated 2 years ago
- [NeurIPS'20] Code for the Paper Compositional Visual Generation and Inference with Energy Based Models☆43Updated last year
- PyTorch implementation for "Training and Inference on Any-Order Autoregressive Models the Right Way", NeurIPS 2022 Oral, TPM 2023 Best Pa…☆11Updated last year
- Simple data balancing baselines for worst-group-accuracy benchmarks.☆39Updated 10 months ago
- ☆17Updated last year
- PyTorch implementation for our ICLR 2024 paper "Diffusion Generative Flow Samplers: Improving learning signals through partial trajectory…☆21Updated 8 months ago
- Euclidean Wasserstein-2 optimal transportation☆43Updated last year
- NeurIPS'23: Energy Discrepancies: A Score-Independent Loss for Energy-Based Models☆11Updated 10 months ago
- ☆23Updated 3 years ago
- ☆21Updated 2 years ago
- The Official PyTorch Implementation of "VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models" (ICLR 2021 spotlight…☆54Updated last year
- PyTorch implementation of the paper "Continuous Wasserstein-2 Barycenter Estimation without Minimax Optimization" (ICLR 2021)☆21Updated 2 years ago
- ☆57Updated last year
- Pytorch code for "Improving Self-Supervised Learning by Characterizing Idealized Representations"☆40Updated last year
- Implementations of orthogonal and semi-orthogonal convolutions in the Fourier domain with applications to adversarial robustness☆41Updated 3 years ago
- Why Do We Need Weight Decay in Modern Deep Learning? [arXiv, Oct 2023]☆41Updated 11 months ago
- ☆34Updated last month
- Diagnosing Vulnerability of Variational Auto-Encoders to Adversarial Attacks☆13Updated 2 years ago
- ☆25Updated last year
- Code for ICLR 2022 Paper, "Controlling Directions Orthogonal to a Classifier"☆34Updated last year
- ☆25Updated last year