• Stars
    star
    153
  • Rank 243,368 (Top 5 %)
  • Language
    Python
  • Created about 3 years ago
  • Updated over 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

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)

Modern Self-Referential Weight Matrix

This is the official repository containing code for the paper:

A Modern Self-Referential Weight Matrix That Learns to Modify Itself (ICML 2022 & NeurIPS 2021 Deep RL Workshop)

An earlier/shorter version of the paper (only containing the RL part) was presented at NeurIPS 2021 Deep RL Workshop. The corresponding version is available on Openreview.

This reposity also contains code for the paper: Accelerating Neural Self-Improvement via Bootstrapping (ICLR 2023 Workshop). Example scripts for this paper can be found under supervised_learning/scripts/bootstrapping.

General instructions

Please refer to the readme file under each directory for further instructions.

License files can be found under the corresponding directories.

In all tasks, our custom CUDA kernels will be automatically compiled. To avoid recompiling the code multiple times, we recommend to specify the path to a directory to store the compiled code via:

export TORCH_EXTENSIONS_DIR="/home/me/torch_extensions/rl"

BibTex

ICML 2022:

@inproceedings{irie2022modern,
  title={A Modern Self-Referential Weight Matrix That Learns to Modify Itself},
  author={Kazuki Irie and Imanol Schlag and R\'obert Csord\'as and J\"urgen Schmidhuber},
  booktitle={Proc. Int. Conf. on Machine Learning (ICML)},
  address={Baltimore, {MD}, {USA}},
  month=jul,
  year={2022}
}

NeurIPS 2021 Workshop:

@inproceedings{irie2021modern,
  title={A Modern Self-Referential Weight Matrix That Learns to Modify Itself}, 
  author={Kazuki Irie and Imanol Schlag and R\'obert Csord\'as and J\"urgen Schmidhuber},
  booktitle={Workshop on Deep Reinforcement Learning, NeurIPS},
  address={Virtual only},
  year={2021}
}

ICLR 2023 Workshop:

@inproceedings{irie2023accelerating,
  title={Accelerating Neural Self-Improvement via Bootstrapping},
  author={Kazuki Irie and J{\"u}rgen Schmidhuber},
      booktitle={Workshop on Mathematical and Empirical Understanding of Foundation Models, ICLR},
      address={Kigali, Rwanda},
      year={2023}
}

Links

More Repositories

1

sacred

Sacred is a tool to help you configure, organize, log and reproduce experiments developed at IDSIA.
Python
4,205
star
2

brainstorm

Fast, flexible and fun neural networks.
Python
1,304
star
3

hhmarl_2D

Heterogeneous Hierarchical Multi Agent Reinforcement Learning for Air Combat
Python
59
star
4

recurrent-fwp

Official repository for the paper "Going Beyond Linear Transformers with Recurrent Fast Weight Programmers" (NeurIPS 2021)
Python
46
star
5

neuraldiffeq-fwp

Official repository for the paper "Neural Differential Equations for Learning to Program Neural Nets Through Continuous Learning Rules" (NeurIPS 2022)
Python
17
star
6

credici

Credici: Credal Inference for Causal Inference
Java
16
star
7

lmtool-fwp

PyTorch Language Modeling Toolkit for Fast Weight Programmers
Python
14
star
8

novel2graph

Python
13
star
9

fpainter

Official repository for the paper "Images as Weight Matrices: Sequential Image Generation Through Synaptic Learning Rules" (ICLR 2023)
Python
12
star
10

GoGePo

Official repository for the paper "Goal-Conditioned Generators of Deep Policies"
Python
10
star
11

crema

Crema: Credal Models Algorithms
Java
10
star
12

gpforecasting

Python
7
star
13

kohonen-vae

Official repository for the paper "Topological Neural Discrete Representation Learning ร  la Kohonen" (ICML 2023 Workshop on Sampling and Optimization in Discrete Space)
Python
5
star
14

policyevaluator

Official repository for the paper "General Policy Evaluation and Improvement by Learning to Identify Few But Crucial States"
Python
5
star
15

bayesRecon

Source of bayesRecon R package ๐Ÿ“ˆ
R
5
star
16

MLprod

Machine Learning in Production
Jupyter Notebook
4
star
17

NewTechnoWar

New Techno War, an IDSIA project in collaboration with Armasuisse.
Jupyter Notebook
3
star
18

crepo

Java
2
star
19

flotta

A federated learning framework for researchers.
Python
2
star
20

rtrl-elstm

Official repository for the paper "Exploring the Promise and Limits of Real-Time Recurrent Learning"
Python
1
star
21

adapquest

ADAPtive QUESTtionnaire, an IDSIA tool for adaptive tests, surveys, and questionnaires.
Java
1
star