​Learning in Evolution, Evolution in Learning
TWCF Number
Project Duration
April 16 / 2018
- April 15 / 2020
Core Funding Area
Big Questions
Amount Awarded

* A Grant DOI (digital object identifier) is a unique, open, global, persistent and machine-actionable identifier for a grant.

Eörs Szathmáry
Institution Parmenides Stiftung

Intelligence is the ability to process external or internal information with the goal of becoming more adaptive in particular contexts. Learning and creativity are key aspects of this capacity. This project is part of a new wave of interest in the relationship between evolution and learning, but it goes beyond earlier work by specifying the shared mechanisms.

We begin from the observation of deep similarities and formal relationships between the two, and examine the idea that there is one unified theory behind the two manifestations. Our project has two major streams. In “Learning in Evolution” (i.e., how learning processes can be effective in evolution), we will elaborate on this link as applied to (i) unsupervised learning in ecosystems, and (ii) reinforcement learning in embryonic development. We also aim to prove the hypothesis that (iii) the deep learning mechanism can help us re-formulate my book The Major Transitions from lower to higher evolutionary units (as passage from lower to higher level features of hierarchical representations). In “Evolution in Learning” (i.e., how evolutionary processes can be effective in learning), we test our hypothesis that Darwinian neurodynamics can provide novel understanding of (i) basic linguistic operations, and (ii) the scope of facilitated variation in generating new ideas.

This project aims to synthesize the two approaches by linking and mapping their respective logical and mathematical foundations. We will use logical–conceptual analysis and computational modeling to tackle these questions. If we are correct, scholars will have an altered view about both evolution and learning, and the field may move closer to hypothesizing one theory encompassing different subcases. We will produce up to six high-profile research papers (roughly divided into the two areas) and one overarching review.

Project Resources
Insight problems are particularly interesting, because problems which require restructuring allow researchers to investigate the underpinnings...
Bayesian learning theory and evolutionary theory both formalize adaptive competition dynamics in possibly high-dimensional, varying, and noisy...
Abstract Our understanding of real-world connected systems has benefited from studying their evolution, from random wirings and rewirings to g...
Efficient search in vast combinatorial spaces, such as those of possible action sequences, linguistic structures, or causal explanations, is a...
There is increased awareness of the possibility of developmental memories resulting from evolutionary learning. Genetic regulatory and neural ...
Abstract The concept of entropy connects the number of possible configurations with the number of variables in large stochastic systems. Indep...
Complexity of life forms on the Earth has increased tremendously, primarily driven by subsequent evolutionary transitions in individuality, a ...
Opinions expressed on this page, or any media linked to it, do not necessarily reflect the views of Templeton World Charity Foundation, Inc. Templeton World Charity Foundation, Inc. does not control the content of external links.
Person doing research
Projects &
Explore the projects we’ve funded. We’ve awarded hundreds of grants to researchers and institutions worldwide.

Projects & Resources