# Meta-Learning by the Baldwin Effect

arXiv: https://arxiv.org/abs/1806.07917 · HTML: https://mlsanity.com/paper/1806.07917/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Jun 2018.

## Built on by (most cited first, 10 of 10)

- [Transformers as Meta-Learners for Implicit Neural Representations](https://mlsanity.com/paper/2208.02801/) (2022)
- [Evolutionary Reinforcement Learning: A Survey](https://mlsanity.com/paper/2303.04150/) (2023)
- [Reset It and Forget It: Relearning Last-Layer Weights Improves Continual and Transfer Learning](https://mlsanity.com/paper/2310.07996/) (2023)
- [Evolutionary Optimization of Physics-Informed Neural Networks: Advancing Generalizability by the Baldwin Effect](https://mlsanity.com/paper/2312.03243/) (2023)
- [Provably Safe Model-Based Meta Reinforcement Learning: An Abstraction-Based Approach](https://mlsanity.com/paper/2109.01255/) (2021)
- [The Price equation reveals a universal force-metric-bias law of algorithmic learning and natural selection](https://mlsanity.com/paper/2507.18549/) (2025)
- [BiERL: A Meta Evolutionary Reinforcement Learning Framework via Bilevel Optimization](https://mlsanity.com/paper/2308.01207/) (2023)
- [Breaching the Bottleneck: Evolutionary Transition from Reward-Driven Learning to Reward-Agnostic Domain-Adapted Learning in Neuromodulated Neural Nets](https://mlsanity.com/paper/2404.12631/) (2024)
- [Decoupling Representation and Learning in Genetic Programming: the LaSER Approach](https://mlsanity.com/paper/2505.17309/) (2025)
- [Distilling a Modular Reservoir Through a Genomic Bottleneck](https://mlsanity.com/paper/2606.28380/) (2026)

## Abstract (arXiv)

The scope of the Baldwin effect was recently called into question by two papers that closely examined the seminal work of Hinton and Nowlan. To this date there has been no demonstration of its necessity in empirically challenging tasks. Here we show that the Baldwin effect is capable of evolving few-shot supervised and reinforcement learning mechanisms, by shaping the hyperparameters and the initial parameters of deep learning algorithms. Furthermore it can genetically accommodate strong learning biases on the same set of problems as a recent machine learning algorithm called MAML "Model Agnostic Meta-Learning" which uses second-order gradients instead of evolution to learn a set of reference parameters (initial weights) that can allow rapid adaptation to tasks sampled from a distribution. Whilst in simple cases MAML is more data efficient than the Baldwin effect, the Baldwin effect is more general in that it does not require gradients to be backpropagated to the reference parameters or hyperparameters, and permits effectively any number of gradient updates in the inner loop. The Baldwin effect learns strong learning dependent biases, rather than purely genetically accommodating fixed behaviours in a learning independent manner.