bgrimmer / LongStepCertificatesLinks
Certificates proving the convergence rates claimed in Table 1 of the (forthcoming) paper "Provably Faster Gradient Descent via Long Steps" by Benjamin Grimmer. The Mathematica notebooks include everything in rational form and computations (exact arithmetic) verifying all of the need (spectral) properties of the certificates.
☆8Updated last year
Alternatives and similar repositories for LongStepCertificates
Users that are interested in LongStepCertificates are comparing it to the libraries listed below
Sorting:
- Repo for solving arc problems with an Neural Cellular Automata☆15Updated 2 weeks ago
- Training hybrid models for dummies.☆21Updated 4 months ago
- code for paper "Accessing higher dimensions for unsupervised word translation"☆21Updated last year
- [ICML 2025] Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction☆23Updated last week
- Minimum Description Length probing for neural network representations☆19Updated 4 months ago
- 🧮 Algebraic Positional Encodings.☆13Updated 4 months ago
- Experimental scripts for researching data adaptive learning rate scheduling.☆23Updated last year
- Exploration into the Firefly algorithm in Pytorch☆39Updated 3 months ago
- ☆18Updated last year
- GoldFinch and other hybrid transformer components☆10Updated 3 weeks ago
- Official code for paper: Conservative objective models are a special kind of contrastive divergence-based energy model☆14Updated last year
- Implementation of a holodeck, written in Pytorch☆18Updated last year
- Implementation of Spectral State Space Models☆16Updated last year
- ☆23Updated 5 months ago
- Implementation and explorations into Blackbox Gradient Sensing (BGS), an evolutionary strategies approach proposed in a Google Deepmind p…☆13Updated last week
- ☆14Updated last month
- Fast singularity detection with kernel☆33Updated last year
- ☆11Updated last year
- We study toy models of skill learning.☆28Updated 4 months ago
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated last year
- ☆13Updated 2 years ago
- DiCE: The Infinitely Differentiable Monte-Carlo Estimator☆31Updated last year
- This library supports evaluating disparities in generated image quality, diversity, and consistency between geographic regions.☆20Updated last year
- High-performance tokenized language data-loader for Python C++ extension☆13Updated 10 months ago
- Code for Arxiv Double Descent Demystified: Identifying, Interpreting & Ablating the Sources of a Deep Learning Puzzle☆26Updated last year
- ☆16Updated last year
- Automatic Integration for Neural Spatio-Temporal Point Process models (AI-STPP) is a new paradigm for exact, efficient, non-parametric inf…☆24Updated 7 months ago
- Automaton & Cognition☆16Updated last year
- Source-to-Source Debuggable Derivatives in Pure Python☆15Updated last year
- Generative Equilibrium Transformer☆18Updated last year