Infinite Hidden Conditional Random Fields for Human Behavior Analysis

• Published in: IEEE transaction on neural networks and learning systems Vol. 24; no. 1; pp. 170 - 177
• Main Authors: Bousmalis, K., Zafeiriou, S., Morency, L., Pantic, M.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.01.2013; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; Behavior; Classification; Computer science; control theory; systems; Computer Simulation; Dirichlet problem; Discriminative models; Exact sciences and technology; hidden conditional random fields; Hidden Markov models; Humans; Information retrieval. Graph; Learning; Learning and adaptive systems; Learning systems; Markov Chains; Mathematical model; Monte Carlo Method; Monte Carlo methods; Monte Carlo simulation; Neural networks; nonparametric Bayesian learning; Pain; Pattern Recognition, Automated; Sampling; Studies; Tasks; Theoretical computing; Training; Trajectory; Vectors; Bayes estimation; Monte Carlo method; Latent variable model; Cross validation; Conditional probability; nonparametric Bayesian learning; Modeling; Dirichlet distribution; Discriminative models; Markov chain; Restaurant analogy; Behavioral analysis; hidden conditional random fields; Classification; Internal structure; Probability learning; Graph method; Sampling; Data structure; Learning algorithm
• ISSN: 2162-237X; 2162-2388
• DOI: 10.1109/TNNLS.2012.2224882

Hidden conditional random fields (HCRFs) are discriminative latent variable models that have been shown to successfully learn the hidden structure of a given classification problem (provided an appropriate validation of the number of hidden states). In this brief, we present the infinite HCRF (iHCRF), which is a nonparametric model based on hierarchical Dirichlet processes and is capable of automatically learning the optimal number of hidden states for a classification task. We show how we learn the model hyperparameters with an effective Markov-chain Monte Carlo sampling technique, and we explain the process that underlines our iHCRF model with the Restaurant Franchise Rating Agencies analogy. We show that the iHCRF is able to converge to a correct number of represented hidden states, and outperforms the best finite HCRFs-chosen via cross-validation-for the difficult tasks of recognizing instances of agreement, disagreement, and pain. Moreover, the iHCRF manages to achieve this performance in significantly less total training, validation, and testing time.