Learning by Abstraction: The Neural State Machine

NeurIPS 2019 Drew A. HudsonChristopher D. Manning

We introduce the Neural State Machine, seeking to bridge the gap between the neural and symbolic views of AI and integrate their complementary strengths for the task of visual reasoning. Given an image, we first predict a probabilistic graph that represents its underlying semantics and serves as a structured world model... (read more)

PDF Abstract NeurIPS 2019 PDF NeurIPS 2019 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Visual Question Answering GQA test-dev NSM Accuracy 62.95 # 1
Visual Question Answering GQA test-std NSM Accuracy 63.17 # 1
Visual Question Answering VQA-CP NSM Score 45.8 # 5

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet