A Modern Self-Referential Weight Matrix That Learns to Modify Itself

J Ürgen Schmidhuber, Kazuki Irie, Imanol Schlag, Róbert Csordás

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 9 functions out of this paper's own repositories and ran 8 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

Repositories linked to this paper

• idsia/modern-srwm — marked official in Papers with Code; named in the paper
• idsia/neuraldiffeq-fwp
• IDSIA/modern-srwm

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The weight matrix (WM) of a neural network (NN) is its program. The programs of many traditional NNs are learned through gradient descent in some error function, then remain fixed. The WM of a self-referential NN, however, can keep rapidly modifying all of itself during runtime. In principle, such NNs can meta-learn to learn, and meta-meta-learn to meta-learn to learn, and so on, in the sense of recursive self-improvement. While NN architectures potentially capable of implementing such behaviour have been proposed since the '90s, there have been few if any practical studies. Here we revisit such NNs, building upon recent successes of fast weight programmers and closely related linear Transformers. We propose a scalable self-referential WM (SRWM) that learns to use outer products and the delta update rule to modify itself. We evaluate our SRWM in supervised few-shot learning and in multi-task reinforcement learning with procedurally generated game environments. Our experiments demonstrate both practical applicability and competitive performance of the proposed SRWM. Our code is public † .

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2202.05780")
get_code_for_paper("2202.05780")
have("2202.05780")

Connect an agent — have() is free.