Gold Seeker: Information Gain from Policy Distributions for Goal-oriented Vision-and-Langauge Reasoning
This addresses the problem of active information acquisition in vision-and-language reasoning for researchers in AI and computer vision, representing an incremental improvement over existing methods.
The paper tackles the challenge of identifying and selecting actions to acquire necessary information for goal-oriented vision-and-language reasoning by proposing a reinforcement learning approach that maintains a distribution over internal information to assess potential information gain. The method outperforms competitors on visual dialog and visual query generation tasks.
As Computer Vision moves from a passive analysis of pixels to active analysis of semantics, the breadth of information algorithms need to reason over has expanded significantly. One of the key challenges in this vein is the ability to identify the information required to make a decision, and select an action that will recover it. We propose a reinforcement-learning approach that maintains a distribution over its internal information, thus explicitly representing the ambiguity in what it knows, and needs to know, towards achieving its goal. Potential actions are then generated according to this distribution. For each potential action a distribution of the expected outcomes is calculated, and the value of the potential information gain assessed. The action taken is that which maximizes the potential information gain. We demonstrate this approach applied to two vision-and-language problems that have attracted significant recent interest, visual dialog and visual query generation. In both cases, the method actively selects actions that will best reduce its internal uncertainty and outperforms its competitors in achieving the goal of the challenge.