RobertCsordas / modulesLinks
The official repository for our paper "Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks". We develop a method for analyzing emerging functional modularity in neural networks based on differentiable weight masks and use it to point out important issues in current-day neural networks.
☆46Updated 2 years ago
Alternatives and similar repositories for modules
Users that are interested in modules are comparing it to the libraries listed below
Sorting:
- Experiments and code to generate the GINC small-scale in-context learning dataset from "An Explanation for In-context Learning as Implici…☆108Updated last year
- ☆34Updated 4 years ago
- ☆65Updated last year
- ☆62Updated 4 years ago
- Official code for "In Search of Robust Measures of Generalization" (NeurIPS 2020)☆28Updated 4 years ago
- An Investigation of Why Overparameterization Exacerbates Spurious Correlations☆30Updated 5 years ago
- ☆55Updated 5 years ago
- Energy-Based Models for Continual Learning Official Repository (PyTorch)☆44Updated 3 years ago
- ☆34Updated 4 years ago
- [NeurIPS 2021] A Geometric Analysis of Neural Collapse with Unconstrained Features☆59Updated 3 years ago
- Group-conditional DRO to alleviate spurious correlations☆15Updated 4 years ago
- Code to implement the AND-mask and geometric mean to do gradient based optimization, from the paper "Learning explanations that are hard …☆41Updated 4 years ago
- ☆46Updated 2 years ago
- Toy datasets to evaluate algorithms for domain generalization and invariance learning.☆42Updated 3 years ago
- [ICLR'22] Self-supervised learning optimally robust representations for domain shift.☆24Updated 3 years ago
- ☆19Updated 3 years ago
- Compositional Explanations of Neurons, NeurIPS 2020 https://arxiv.org/abs/2006.14032☆25Updated 4 years ago
- ☆81Updated last year
- ☆62Updated 3 years ago
- Code to reproduce the results for Compositional Attention☆59Updated 2 years ago
- Gradient Starvation: A Learning Proclivity in Neural Networks☆61Updated 4 years ago
- "Predict, then Interpolate: A Simple Algorithm to Learn Stable Classifiers" ICML 2021☆18Updated 4 years ago
- Code accompanying paper: Meta-Learning to Improve Pre-Training☆37Updated 3 years ago
- Contains code for the NeurIPS 2020 paper by Pan et al., "Continual Deep Learning by FunctionalRegularisation of Memorable Past"☆44Updated 4 years ago
- Simple data balancing baselines for worst-group-accuracy benchmarks.☆42Updated last year
- Code for "Supermasks in Superposition"☆124Updated 2 years ago
- ModelDiff: A Framework for Comparing Learning Algorithms☆59Updated 2 years ago
- ☆24Updated 4 years ago
- Code for "Just Train Twice: Improving Group Robustness without Training Group Information"☆72Updated last year
- ☆85Updated last year