DDRprog: A CLEVR Differentiable Dynamic Reasoning Programmer
This work addresses a bottleneck in dynamic reasoning for AI systems, offering incremental improvements in data efficiency and structural supervision for specific domains.
The paper tackles the problem of nondifferentiability in dynamic architectures by introducing a Dynamic Differentiable Reasoning (DDR) framework, which learns branching programs and their functions jointly, achieving improvements in subtask consistency and generalization to long expressions in tasks like CLEVR Visual Question Answering and reverse Polish notation evaluation.
We present a novel Dynamic Differentiable Reasoning (DDR) framework for jointly learning branching programs and the functions composing them; this resolves a significant nondifferentiability inhibiting recent dynamic architectures. We apply our framework to two settings in two highly compact and data efficient architectures: DDRprog for CLEVR Visual Question Answering and DDRstack for reverse Polish notation expression evaluation. DDRprog uses a recurrent controller to jointly predict and execute modular neural programs that directly correspond to the underlying question logic; it explicitly forks subprocesses to handle logical branching. By effectively leveraging additional structural supervision, we achieve a large improvement over previous approaches in subtask consistency and a small improvement in overall accuracy. We further demonstrate the benefits of structural supervision in the RPN setting: the inclusion of a stack assumption in DDRstack allows our approach to generalize to long expressions where an LSTM fails the task.