IDSIA/modern-srwm
Official repository for the paper "A Modern Self-Referential Weight Matrix That Learns to Modify Itself" (ICML 2022 & NeurIPS 2021 Deep RL Workshop) and "Accelerating Neural Self-Improvement via Bootstrapping" (ICLR 2023 Workshop)
This project provides advanced techniques for machine learning researchers and practitioners who are experimenting with novel neural network architectures. It offers methods to build neural networks that can dynamically adjust their own internal connections (weights) as they learn, leading to more efficient and self-improving models. The intended user is a machine learning scientist or deep learning engineer focused on model development and optimization.
176 stars. No commits in the last 6 months.
Use this if you are a researcher or advanced practitioner designing and testing next-generation neural networks, especially in deep reinforcement learning or supervised learning scenarios where models need to adapt and improve over time.
Not ideal if you are looking for an off-the-shelf solution for a standard machine learning task or if you are not comfortable working with experimental deep learning architectures and custom code.
Stars
176
Forks
18
Language
Python
License
—
Category
Last pushed
Jun 11, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/IDSIA/modern-srwm"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
galilai-group/stable-pretraining
Reliable, minimal and scalable library for pretraining foundation and world models
CognitiveAISystems/MAPF-GPT
[AAAI-2025] This repository contains MAPF-GPT, a deep learning-based model for solving MAPF...
UKPLab/gpl
Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled...
larslorch/avici
Amortized Inference for Causal Structure Learning, NeurIPS 2022
svdrecbd/mhc-mlx
MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.