Inferning 2012: ICML Workshop on interaction between Inference and Learning
TechTalks from event: Inferning 2012: ICML Workshop on interaction between Inference and Learning
-
Keynote: Marginalization-Based Parameter Learning in Graphical ModelsLikelihood-based learning of graphical models faces challenges of computational complexity and robustness to model error. This talk will discuss methods that directly maximize a measure of the accuracy of predicted marginals, in the context of a particular approximate inference algorithm. Experiments suggest that marginalization-based learning, by compensating for both model and inference approximations at training time, can perform better than likelihood-based approximations when the model being fit is approximate in nature.
- All Talks
- Opening Remarks
- Keynote: Learning Tractable but Expressive Models
- On the Mismatch Between Learning and Inference for Single Network Domains
- A Framework for Tuning Posterior Entropy in Unsupervised Learning
- Keynote: Modeling, Inference and Estimation using Random MAP Perturbations
- Cost-sensitive Dynamic Feature Selection
- Speeding up MAP with Column Generation and Block Regularization
- Learning Search Based Inference for Object Detection
- Fast and Accurate Prediction via Evidence-Specific MRF Structure
- Keynote: Learning Approximate Inference Policies for Fast Prediction
- Approximating Marginals Using Discrete Energy Minimization
- Keynote: Marginalization-Based Parameter Learning in Graphical Models
- Learned Prioritization for Trading Off Accuracy and Speed