Bayesian Recommender Systems
Recommender systems are increasingly driving user experiences on the Internet. This personalization is often achieved through the factorization of a large but sparse observation matrix of user-item feedback signals. We present a probabilistic model for generating personalised recommendations of items to users. The system makes use of content information in the form of user and item meta data in combination with collaborative filtering information from previous user behavior in order to predict the value of an item for a user. Users and items are represented by feature vectors which are mapped into a low-dimensional trait space in which similarity is measured in terms of inner products. The model can be trained from different types of feedback in order to learn user-item preferences. Efficient inference is achieved by approximate message passing involving a combination of Expectation Propagation (EP) and Variational Message Passing. By using Assumed-Density Filtering (ADF) for training, the model requires only a single pass through the training data.
We also include a dynamics model which allows an item’s popularity, a user’s taste or a user’s personal rating scale to drift over time. This is an on-line learning algorithm capable of incrementally taking account of new data so the system can immediately reflect the latest user preferences. In instances where the user’s social network is known, its inclusion can significantly improve recommendations for cold start users. We also propose and investigate two ways for including a social network, either as a Markov Random Field that describes a user similarity in the prior over user features, or an explicit model that treats social links as observations.
A Bayesian Treatment of Social Links in Recommender Systems Technical Report
CU Technical Report CU-CS-1092-12 2012.
Matchbox: Large Scale Online Bayesian Recommendations Proceedings Article
In: Proceedings of the 18th International Conference on World Wide Web, pp. 111–120, 2009.
Large-Scale Approximate Recommender Systems
In general, collaborative filtering (CF) systems show their real strength when supplied with enormous data sets. In order to handle massive amounts of information, we investigated sketching techniques. We demonstrate how to use fingerprinting methods to compute a family of rank correlation coefficients with high accuracy and confidence. We examine the suggested methods empirically through a recommender system for the Netflix dataset, showing that the required fingerprint sizes are even smaller than the theoretical analysis suggests. We also explore the of use standard hash functions rather than min-wise independent hashes and the relation between the quality of the final recommendations and the fingerprint size.
In: Proceedings of 17th International Symposium on String Processing and Information Retrieval, pp. 25–36, 2010.
In: Proceedings of 16th International Symposium String Processing and Information Retrieval, pp. 344–352, 2009.
Active Learning in Recommender Systems
Recommendation systems are trained to predict user behavior which is then used to generate new training by receiving actual user behavior of these recommendations. This creates a feedback loop: as long as the low-cost way to interact with the service is through the recommender, the recommender will only ever see behavioral data on the items it chooses. This process can lead to hidden biases, as it effectively limits how much information the recommender system will ever see. We explore the notion that recommender systems are a special kind of active learning agents, with the peculiarity that the cost of asking for the label of an instance depends on its true label, as the cost of showing a bad recommendation when exploring is higher than the cost of showing a good recommendation.
A Penny for Your Thoughts? The Value of Information in Recommendation Systems Proceedings Article
In: NIPS Workshop on Bayesian Optimization, Experimental Design, and Bandits, pp. 9–14, 2011.
Transparent User Models
Personalization is a ubiquitous phenomenon in our daily online experience. While such technology is critical for helping us combat the overload of information we face, in many cases, we may not even realize that our results are being tailored to our personal tastes and preferences. Worse yet, when such a system makes a mistake, we have little recourse to correct it. In this work, we developed a new user-interpretable feature set upon which to base personalized recommendations. These features, which we call badges, represent fundamental traits of users (e.g., “vegetarian” or “Apple fanboy”) inferred by modeling the interplay between a user’s behavior and self-reported identity.
Transparent User Models for Personalization Proceedings Article
In: Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 678–686, ACM 2012.
Collaborative Expert Portfolio Management
We consider the task of assigning experts from a portfolio of specialists in order to solve a set of tasks by combining collaborative filtering with a feature-based description of tasks and experts to yield a general framework for managing a portfolio of experts. The model learns an embedding of tasks and problems into a latent space in which affinity is measured by the inner product. The model can be trained incrementally and can track non-stationary data, tracking potentially changing expert and task characteristics. The approach allows us to use a principled decision theoretic framework for expert selection and recommendation, allowing the user to choose a utility function that best suits their objectives. We apply the model to manage a portfolio of algorithms to solve hard combinatorial problems. This is a well studied area and we demonstrate a large improvement on the state of the art in one domain (constraint solving) and in a second domain (combinatorial auctions) created a portfolio that performed significantly better than any single algorithm.