Multi-Task Optimization over Networks of Tasks

TL;DR

MONET is a novel multi-task optimization algorithm that models task relationships as a graph, enabling efficient knowledge transfer and scalability to large task sets in high-dimensional problems.

Contribution

It introduces a graph-based task modeling approach that improves scalability and topology awareness in multi-task optimization, surpassing existing MAP-Elites variants.

Findings

MONET matches or exceeds MAP-Elites baselines across four diverse domains.

It effectively scales to thousands of tasks in high-dimensional spaces.

The graph-based approach exploits task topology for better knowledge transfer.

Abstract

Multi-task optimization is a powerful approach for solving a large number of tasks in parallel. However, existing algorithms face distinct limitations: Population-based methods scale poorly and remain underexplored for large task sets. Approaches that do scale beyond a thousand tasks are mostly MAP-Elites variants and rely on a fixed, discretized archive that disregards the topology of the task space. We introduce MONET (Multi-Task Optimization over Networks of Tasks), a multi-task optimization algorithm that models the task space as a graph: tasks are nodes, and edges connect tasks in the task parameter space. This representation enables knowledge transfer between tasks and remains tractable for high-dimensional problems while exploiting the topology of the task space. MONET combines social learning, which generates candidates from neighboring nodes via crossover, with individual learning, which refines a node's own solution independently via mutation. We evaluate MONET on four domains (archery, arm, and cartpole with 5,000 tasks each; hexapod with 2,000 tasks) and show that it matches or exceeds the performance of existing MAP-Elites-based baselines across all four domains.