Meta-Learning by the Baldwin Effect

Abstract

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.

References in corpus (9)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Matching Networks for One Shot Learning
• RL${}^2$: Fast Reinforcement Learning via Slow Reinforcement Learning
• One-shot Learning with Memory-Augmented Neural Networks
• Population Based Training of Neural Networks
• Myths and Legends of the Baldwin Effect
• Phenotypic Plasticity, the Baldwin Effect, and the Speeding up of Evolution: the Computational Roots of an Illusion
• Lamarckian Evolution and the Baldwin Effect in Evolutionary Neural Networks
• The revival of the Baldwin Effect

Cited by in corpus (11)

• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
• ES-MAML: Simple Hessian-Free Meta Learning
• Meta-Learning through Hebbian Plasticity in Random Networks
• Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning
• Meta-learnt priors slow down catastrophic forgetting in neural networks
• Evolving Self-supervised Neural Networks: Autonomous Intelligence from Evolved Self-teaching
• Population-Based Evolution Optimizes a Meta-Learning Objective
• Neuro-evolutionary Frameworks for Generalized Learning Agents
• Why to "grow" and "harvest" deep learning models?
• On the potential for open-endedness in neural networks
• Meta Reinforcement Learning with Distribution of Exploration Parameters Learned by Evolution Strategies