RL policies decompose into information-regularized primitives that compete by requesting state information amounts, with the greediest one acting, yielding better generalization than flat or hierarchical baselines.
Routing Networks and the Challenges of Modular and Compositional Computation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Compositionality is a key strategy for addressing combinatorial complexity and the curse of dimensionality. Recent work has shown that compositional solutions can be learned and offer substantial gains across a variety of domains, including multi-task learning, language modeling, visual question answering, machine comprehension, and others. However, such models present unique challenges during training when both the module parameters and their composition must be learned jointly. In this paper, we identify several of these issues and analyze their underlying causes. Our discussion focuses on routing networks, a general approach to this problem, and examines empirically the interplay of these challenges and a variety of design decisions. In particular, we consider the effect of how the algorithm decides on module composition, how the algorithm updates the modules, and if the algorithm uses regularization.
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Reinforcement Learning with Competitive Ensembles of Information-Constrained Primitives
RL policies decompose into information-regularized primitives that compete by requesting state information amounts, with the greediest one acting, yielding better generalization than flat or hierarchical baselines.