Learning One Abstract Bit at a Time Through Self-invented Experiments Encoded as Neural Networks

Vincent Herrmann*, Louis Kirsch, Jürgen Schmidhuber

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


There are two important things in science: (A) Finding answers to given questions, and (B) Coming up with good questions. Our artificial scientists not only learn to answer given questions, but also continually invent new questions, by proposing hypotheses to be verified or falsified through potentially complex and time-consuming experiments, including thought experiments akin to those of mathematicians. While an artificial scientist expands its knowledge, it remains biased towards the simplest, least costly experiments that still have surprising outcomes, until they become boring. We present an empirical analysis of the automatic generation of interesting experiments. In the first setting, we investigate self-invented experiments in a reinforcement-providing environment and show that they lead to effective exploration. In the second setting, pure thought experiments are implemented as the weights of recurrent neural networks generated by a neural experiment generator. Initially interesting thought experiments may become boring over time.

Original languageEnglish (US)
Title of host publicationActive Inference - 4th International Workshop, IWAI 2023, Revised Selected Papers
EditorsChristopher L. Buckley, Daniela Cialfi, Pablo Lanillos, Maxwell Ramstead, Tim Verbelen, Maxwell Ramstead, Noor Sajid, Hideaki Shimazaki, Martijn Wisse
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages21
ISBN (Print)9783031479571
StatePublished - 2024
Event4th International Workshop on Active Inference, IWAI 2023 - Ghent, Belgium
Duration: Sep 13 2023Sep 15 2023

Publication series

NameCommunications in Computer and Information Science
Volume1915 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937


Conference4th International Workshop on Active Inference, IWAI 2023

Bibliographical note

Publisher Copyright:
© 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.


  • Exploration
  • Reinforcement Learning

ASJC Scopus subject areas

  • General Computer Science
  • General Mathematics


Dive into the research topics of 'Learning One Abstract Bit at a Time Through Self-invented Experiments Encoded as Neural Networks'. Together they form a unique fingerprint.

Cite this