Timezone: »

ZeroC: A Neuro-Symbolic Model for Zero-shot Concept Recognition and Acquisition at Inference Time
Tailin Wu · Megan Tjandrasuwita · Zhengxuan Wu · Xuelin Yang · Kevin Liu · Rok Sosic · Jure Leskovec

Wed Nov 30 02:00 PM -- 04:00 PM (PST) @ Hall J #625

Humans have the remarkable ability to recognize and acquire novel visual concepts in a zero-shot manner. Given a high-level, symbolic description of a novel concept in terms of previously learned visual concepts and their relations, humans can recognize novel concepts without seeing any examples. Moreover, they can acquire new concepts by parsing and communicating symbolic structures using learned visual concepts and relations. Endowing these capabilities in machines is pivotal in improving their generalization capability at inference time. In this work, we introduce Zero-shot Concept Recognition and Acquisition (ZeroC), a neuro-symbolic architecture that can recognize and acquire novel concepts in a zero-shot way. ZeroC represents concepts as graphs of constituent concept models (as nodes) and their relations (as edges). To allow inference time composition, we employ energy-based models (EBMs) to model concepts and relations. We design ZeroC architecture so that it allows a one-to-one mapping between a symbolic graph structure of a concept and its corresponding EBM, which for the first time, allows acquiring new concepts, communicating its graph structure, and applying it to classification and detection tasks (even across domains) at inference time. We introduce algorithms for learning and inference with ZeroC. We evaluate ZeroC on a challenging grid-world dataset which is designed to probe zero-shot concept recognition and acquisition, and demonstrate its capability.

Author Information

Tailin Wu (Stanford)
Tailin Wu

Tailin Wu is a postdoc researcher in the Department of Computer Science at Stanford University, working with professor Jure Leskovec. His research interests lies in AI for large-scale simulations of complex systems, AI for scientific discovery, and representation learning

Megan Tjandrasuwita (Massachusetts Institute of Technology)
Zhengxuan Wu (Stanford University)
Xuelin Yang (Stanford University)
Kevin Liu (Stanford University)
Kevin Liu

Student at Stanford University, with an interest in entrepreneurship, trustworthy AI, and systems engineering.

Rok Sosic (Computer Science Department, Stanford University)
Jure Leskovec (Stanford University/Pinterest)

More from the Same Authors