Multi-Task Structural Learning using Local Task Similarity induced Neuron Creation and Removal

Multi-task learning has the potential to improve generalization by maximizing positive transfer between tasks while reducing task interference. Fully achieving this potential is hindered by manually designed architectures that remain static throughout training. In contrast, learning in the brain occurs through structural changes that are in tandem with changes in synaptic strength. Therefore, we propose Multi-Task Structural Learning (MTSL) which simultaneously learns the multi-task architecture and its parameters. MTSL begins with an identical single task network for each task and alternates between a task learning phase and a structural learning phase. In the task learning phase, each network specializes in the corresponding task. In each of the structural learning phases, starting from the earliest layer, locally similar task layers first transfer their knowledge to a newly created group layer after which they become redundant and are removed. Our experimental results show that MTSL achieves competitive generalization with various baselines and improves robustness to out-of-distribution data.