Deep Non-Monotonic Reasoning for Visual Abstract Reasoning Tasks
This addresses the challenge of matching human-like reasoning in AI for visual tasks, though it appears incremental as it builds on existing deep learning methods for a specific dataset.
The paper tackles the problem of deep learning models struggling with visual abstract reasoning tasks due to monotonic reasoning, proposing a non-monotonic computational approach that improves effectiveness over existing models on a difficult variant of the RAVEN dataset.
While achieving unmatched performance on many well-defined tasks, deep learning models have also been used to solve visual abstract reasoning tasks, which are relatively less well-defined, and have been widely used to measure human intelligence. However, current deep models struggle to match human abilities to solve such tasks with minimum data but maximum generalization. One limitation is that current deep learning models work in a monotonic way, i.e., treating different parts of the input in essentially fixed orderings, whereas people repeatedly observe and reason about the different parts of the visual stimuli until the reasoning process converges to a consistent conclusion, i.e., non-monotonic reasoning. This paper proposes a non-monotonic computational approach to solve visual abstract reasoning tasks. In particular, we implemented a deep learning model using this approach and tested it on the RAVEN dataset -- a dataset inspired by the Raven's Progressive Matrices test. Results show that the proposed approach is more effective than existing monotonic deep learning models, under strict experimental settings that represent a difficult variant of the RAVEN dataset problem.