Tensor factorization methods provide a useful way to extract latent factors from complex multirelational data, and also for predicting missing data. Developing tensor factorization methods for massive tensors, especially when the data are binary- or count-valued (which is true of most real-world tensors), however, remains a challenge. We develop a scalable probabilistic tensor factorization framework that enables us to perform efficient factorization of massive binary and count tensor data. The framework is based on (i) the Polya-Gamma augmentation strategy which makes the model fully locally conjugate and allows closed-form parameter updates when data are binary- or count-valued; and (ii) an efficient online Expectation Maximization algorithm, which allows processing data in small mini-batches, and facilitates handling massive tensor data. Moreover, various types of constraints on the factor matrices (e.g., sparsity, non-negativity) can be incorporated under the proposed framework, providing good interpretability, which can be useful for qualitative analyses of the results. We apply the proposed framework on analyzing several binary- and count-valued real-world data sets.

We introduce a novel kernel that upgrades the Weisfeiler-Lehman and other graph kernels to effectively exploit high-dimensional and continuous vertex attributes. Graphs are first decomposed into subgraphs. Vertices of the subgraphs are then compared by a kernel that combines the similarity of their labels and the similarity of their structural role, using a suitable vertex invariant. By changing this invariant we obtain a family of graph kernels which includes generalizations of Weisfeiler-Lehman, NSPDK, and propagation kernels. We demonstrate empirically that these kernels obtain state-of-the-art results on relational data sets.

In online display advertising, state-of-the-art Click Through Rate(CTR) prediction algorithms rely heavily on historical information, and they work poorly on growing number of new ads without any historical information. This is known as the the cold start problem. For image ads, current state-of-the-art systems use handcrafted image features such as multimedia features and SIFT features to capture the attractiveness of ads. However, these handcrafted features are task dependent, inflexible and heuristic. In order to tackle the cold start problem in image display ads, we propose a new feature learning architecture to learn the most discriminative image features directly from raw pixels and user feedback in the target task. The proposed method is flexible and does not depend on human heuristic. Extensive experiments on a real world dataset with 47 billion records show that our feature learning method outperforms existing handcrafted features significantly, and it can extract discriminative and meaningful features.

In this paper, we study Bayesian techniques for entity discovery and temporal segmentation of videos. Existing temporal video segmentation techniques are based on low-level features, and are usually suitable for discovering short, homogeneous shots rather than diverse scenes, each of which contains several such shots. We define scenes in terms of semantic entities (eg. persons). This is the first attempt at entity-driven scene discovery in videos, without using meta-data like scripts. The problem is hard because we have no explicit prior information about the entities and the scenes. However such sequential data exhibit temporal coherence in multiple ways, and this provides implicit cues. To capture these, we propose a Bayesian generative model- EntScene, that represents entities with mixture components and scenes with discrete distributions over these components. The most challenging part of this approach is the inference, as it involves complex interactions of latent variables. To this end, we propose an algorithm based on Dynamic Blocked Gibbs Sampling, that attempts to jointly learn the components and the segmentation, by progressively merging an initial set of short segments. The proposed algorithm compares favourably against suitably designed baselines on several TV-series videos. We extend the method to an unexplored problem: temporal co-segmentation of videos containing same entities.

Science ultimately seeks to reliably predict aspects of the future; but, how is this even possible in light of the logical paradox that making a prediction may cause the world to evolve in a manner that defeats it? We show how learning can naturally resolve this conundrum. The problem is studied within a causal or temporal version of the Probably Approximately Correct semantics, extended so that a learner's predictions are first recorded in the states upon which the learned hypothesis is later applied. On the negative side, we make concrete the intuitive impossibility of predicting reliably, even under very weak assumptions. On the positive side, we identify conditions under which a generic learning schema, akin to randomized trials, supports agnostic learnability.

Traditional methods which the features form cluster structures, as can be seen in often rely on R.I.P or its relaxed variants. However, many machine learning [Lehiste, 1976] and computer vision in real applications, features are often correlated problems [Lan et al., 2013; Lowe, 2004]. Due to the fact that to each other, which makes these assumptions many features extractors are similar to each others and they too strong to be useful. In this paper, we reflect the characteristics of the same image, vision features study the sparse recovery problem in which the feature are often correlated and have cluster structures. This correlation matrix is strictly non-R.I.P.. We prove that is even stronger in those systems that have thousands when features exhibit cluster structures, which often to millions of features [Lan et al., 2013; Gan et al., 2015a; happens in real applications, we are able to recover 2015b].

Given several tasks, multi-task learning (MTL) learns multiple tasks jointly by exploring the interdependence between them. The basic assumption in MTL is that those tasks are indeed related. Existing MTL methods model the task relatedness/interdependence in two different ways, either common parameter-sharing or common feature-sharing across tasks. In this paper, we propose a novel multi-task learning method to jointly learn shared parameters and shared feature representation. Our objective is to learn a set of common features with which the tasks are related as closely as possible, therefore common parameters shared across tasks can be optimally learned. We present a detailed deviation of our multi-task learning method and propose an alternating algorithm to solve the non-convex optimization problem. We further present a theoretical bound which directly demonstrates that the proposed multi-task learning method can successfully model the relatedness via joint common parameter- and common feature-learning. Extensive experiments are conducted on several real world multi-task learning datasets. All results demonstrate the effectiveness of our multi-task model and feature joint learning method.

We introduce a framework that enables an agent to autonomously learn its own symbolic representation of a low-level, continuous environment. Propositional symbols are formalized as names for probability distributions, providing a natural means of dealing with uncertain representations and probabilistic plans. We determine the symbols that are sufficient for computing the probability with which a plan will succeed, and demonstrate the acquisition of a symbolic representation in a computer game domain.

A fundamental problem underlying location-based tasks is to construct a complete profile of users' spatiotemporal patterns. In many real-world settings, the sparsity of location data makes it difficult to construct such a profile. As a remedy, we describe a Bayesian probabilistic graphical model, called Collaborative Place Model (CPM), which infers similarities across users to construct complete and time-dependent profiles of users' whereabouts from unsupervised location data. We apply CPM to both sparse and dense datasets, and demonstrate how it both improves location prediction performance and provides new insights into users' spatiotemporal patterns.

Cross-validation (CV) is one of the main tools for performance estimation and parameter tuning in machine learning. The general recipe for computing CV estimate is to run a learning algorithm separately for each CV fold, a computationally expensive process. In this paper, we propose a new approach to reduce the computational burden of CV-based performance estimation. As opposed to all previous attempts, which are specific to a particular learning model or problem domain, we propose a general method applicable to a large class of incremental learning algorithms, which are uniquely fitted to big data problems. In particular, our method applies to a wide range of supervised and unsupervised learning tasks with different performance criteria, as long as the base learning algorithm is incremental. We show that the running time of the algorithm scales logarithmically, rather than linearly, in the number of CV folds. Furthermore, the algorithm has favorable properties for parallel and distributed implementation. Experiments with state-of-the-art incremental learning algorithms confirm the practicality of the proposed method.