Timezone: »

Embedding Inference for Structured Multilabel Prediction
Farzaneh Mirzazadeh · Siamak Ravanbakhsh · Nan Ding · Dale Schuurmans

Mon Dec 07 04:00 PM -- 08:59 PM (PST) @ 210 C #43

A key bottleneck in structured output prediction is the need for inference during training and testing, usually requiring some form of dynamic programming. Rather than using approximate inference or tailoring a specialized inference method for a particular structure---standard responses to the scaling challenge---we propose to embed prediction constraints directly into the learned representation. By eliminating the need for explicit inference a more scalable approach to structured output prediction can be achieved, particularly at test time. We demonstrate the idea for multi-label prediction under subsumption and mutual exclusion constraints, where a relationship to maximum margin structured output prediction can be established. Experiments demonstrate that the benefits of structured output training can still be realized even after inference has been eliminated.

Author Information

Farzaneh Mirzazadeh (University of Alberta)
Siamak Ravanbakhsh (University of Alberta)
Nan Ding (Google)
Dale Schuurmans (Alberta)

More from the Same Authors