English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper addresses the general problem of reinforcement learning (RL) inpartially observable environments. In 2013, our large RL recurrent neuralnetworks (RNNs) learned from scratch to drive simulated cars fromhigh-dimensional video input. However, real brains are more powerful in manyways. In particular, they learn a predictive model of their initially unknownenvironment, and somehow use it for abstract (e.g., hierarchical) planning andreasoning. Guided by algorithmic information theory, we describe RNN-based AIs(RNNAIs) designed to do the same. Such an RNNAI can be trained on never-endingsequences of tasks, some of them provided by the user, others invented by theRNNAI itself in a curious, playful fashion, to improve its RNN-based worldmodel. Unlike our previous model-building RNN-based RL machines dating back to1990, the RNNAI learns to actively query its model for abstract reasoning andplanning and decision making, essentially "learning to think." The basic ideasof this report can be applied to many other cases where one RNN-like systemexploits the algorithmic information content of another. They are taken from agrant proposal submitted in Fall 2014, and also explain concepts such as"mirror neurons." Experimental results will be described in separate papers.

On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models