Hierarchical Expert Networks for Meta-Learning
arXiv:1911.00348
Abstract
The goal of meta-learning is to train a model on a variety of learning tasks, such that it can adapt to new problems within only a few iterations. Here we propose a principled information-theoretic model that optimally partitions the underlying problem space such that specialized expert decision-makers solve the resulting sub-problems. To drive this specialization we impose the same kind of information processing constraints both on the partitioning and the expert decision-makers. We argue that this specialization leads to efficient adaptation to new tasks. To demonstrate the generality of our approach we evaluate three meta-learning domains: image classification, regression, and reinforcement learning.
arXiv commentPresented at the 4th ICML Workshop on Life Long Machine Learning, 2020
References in corpus (19)
- Deep Learning in Neural Networks: An Overview
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- A Brief Survey of Deep Reinforcement Learning
- Trust Region Policy Optimization
- High-Dimensional Continuous Control Using Generalized Advantage Estimation
- Theoretical Models of Learning to Learn
- Reinforcement Learning with Deep Energy-Based Policies
- Regularizing Neural Networks by Penalizing Confident Output Distributions
- Thermodynamics as a theory of decision-making with information processing costs
- Hierarchically Structured Meta-learning
- Information driven self-organization of complex robotic behaviors
- Bounded rational decision-making from elementary computations that reduce uncertainty
- Meta-learning of Sequential Strategies
- An Information-Theoretic Optimality Principle for Deep Reinforcement Learning
- Scalable Centralized Deep Multi-Agent Reinforcement Learning via Policy Gradients
- Bounded Rational Decision-Making with Adaptive Neural Network Priors
- Divide-and-Conquer Reinforcement Learning
- Options as responses: Grounding behavioural hierarchies in multi-agent RL