We show that without using any domain knowledge, we can predict the final performance of a team of voting agents, at any step towards solving a complex problem.

We propose revisions to the research agenda in Automated Planning. The proposal is based on a review of the role of the Planning Domain Definition Language (PDDL) in the activities of the AI planning community and the impact of PDDL on parts of its research agenda. We specifically show how specific properties of PDDL have impacted research on planning, by putting emphasis on certain research topics and complicating others. We argue that the development of more advanced modeling languages would be — analogously to the impact PDDL has had — a low overhead and smooth route for the ICAPS community shift its research focus to increasingly promising and relevant research topics.

As an important and challenging problem in computer vision and graphics, keypoint-based object tracking is typically formulated in a spatio-temporal statistical learning framework. However, most existing keypoint trackers are incapable of effectively modeling and balancing the following three aspects in a simultaneous manner: temporal model coherence across frames, spatial model consistency within frames, and discriminative feature construction. To address this issue, we propose a robust keypoint tracker based on spatio-temporal multi-task structured output optimization driven by discriminative metric learning. Consequently, temporal model coherence is characterized by multi-task structured keypoint model learning over several adjacent frames, while spatial model consistency is modeled by solving a geometric verification based structured learning problem. Discriminative feature construction is enabled by metric learning to ensure the intra-class compactness and inter-class separability. Finally, the above three modules are simultaneously optimized in a joint learning scheme. Experimental results have demonstrated the effectiveness of our tracker.

Face verification remains a challenging problem in very complex conditions with large variations such as pose, illumination, expression, and occlusions. This problemis exacerbated when we rely unrealistically on a singletraining data source, which is often insufficient to coverthe intrinsically complex face variations. This paperproposes a principled multi-task learning approachbased on Discriminative Gaussian Process Latent VariableModel (DGPLVM), named GaussianFace, for faceverification. In contrast to relying unrealistically on asingle training data source, our model exploits additional data from multiple source-domains to improve the generalization performance of face verification inan unknown target-domain. Importantly, our model can adapt automatically to complex data distributions, and therefore can well capture complex face variations inherent in multiple sources. To enhance discriminative power, we introduced a more efficient equivalent form of Kernel Fisher Discriminant Analysis to DGPLVM.To speed up the process of inference and prediction, we exploited the low rank approximation method. Extensive experiments demonstrated the effectiveness of the proposed model in learning from diverse data sources and generalizing to unseen domains. Specifically, the accuracy of our algorithm achieved an impressive accuracyrate of 98.52% on the well-known and challenging Labeled Faces in the Wild (LFW) benchmark. For the first time, the human-level performance in face verification (97.53%) on LFW is surpassed.

Automatically recognizing a large number of action categories from videos is of significant importance for video understanding. Most existing works focused on the design of more discriminative feature representation, and have achieved promising results when the positive samples are enough. However, very limited efforts were spent on recognizing a novel action without any positive exemplars, which is often the case in the real settings due to the large amount of action classes and the users' queries dramatic variations. To address this issue, we propose to perform action recognition when no positive exemplars of that class are provided, which is often known as the zero-shot learning. Different from other zero-shot learning approaches, which exploit attributes as the intermediate layer for the knowledge transfer, our main contribution is SIR, which directly leverages the semantic inter-class relationships between the known and unknown actions followed by label transfer learning. The inter-class semantic relationships are automatically measured by continuous word vectors, which learned by the skip-gram model using the large-scale text corpus. Extensive experiments on the UCF101 dataset validate the superiority of our method over fully-supervised approaches using few positive exemplars.

We present an unsupervised approach for abnormal event detection in videos. We propose, given a dictionary of features learned from local spatiotemporal cuboids using the sparse coding objective, the abnormality of an event depends jointly on two factors: the frequency of each feature in reconstructing all events (or, rarity of a feature) and the strength by which it is used in reconstructing the current event (or, the absolute coefficient). The Incremental Coding Length (ICL) of a feature is a measure of its entropy gain. Given a dictionary, the ICL computation does not involve any parameter, is computationally efficient and has been used for saliency detection in images with impressive results. In this paper, the rarity of a dictionary feature is learned online as its average energy, a function of its ICL. The proposed approach is applicable to real world streaming videos. Experiments on three benchmark datasets and evaluations in comparison with a number of mainstream algorithms show that the approach is comparable to the state-of-the-art.

This paper is concerned with the problem of domain adaptation with multiple sources from a causal point of view. In particular, we use causal models to represent the relationship between the features X and class label Y , and consider possible situations where different modules of the causal model change with the domain. In each situation, we investigate what knowledge is appropriate to transfer and find the optimal target-domain hypothesis. This gives an intuitive interpretation of the assumptions underlying certain previous methods and motivates new ones. We finally focus on the case where Y is the cause for X with changing PY and PX|Y , that is, PY and PX|Y change independently across domains. Under appropriate assumptions, the availability of multiple source domains allows a natural way to reconstruct the conditional distribution on the target domain; we propose to model PX|Y (the process to generate effect X from cause Y ) on the target domain as a linear mixture of those on source domains, and estimate all involved parameters by matching the target-domain feature distribution. Experimental results on both synthetic and real-world data verify our theoretical results.

If we know most of Smith’s friends are from Boston, what can we say about the rest of Smith’s friends? In this paper, we focus on the node classification problem on networks, which is one of the most important topics in AI and Web communities. Our proposed algorithm which is referred to as OMNIProp has the following properties: (a) seamless and accurate; it works well on any label correlations (i.e., homophily, heterophily, and mixture of them) (b) fast; it is efficient and guaranteed to converge on arbitrary graphs (c) quasi-parameter free; it has just one well-interpretable parameter with heuristic default value of 1. We also prove the theoretical connections of our algorithm to the semi-supervised learning (SSL) algorithms and to random-walks. Experiments on four real, different network datasets demonstrate the benefits of the proposed algorithm, where OMNI-Prop outperforms the top competitors.

While if kernels are not Kernel methods (Schölkopf and Smola 2002; Xu et al. 2009) low rank, Nyström approximations can usually lead to suboptimal have received a lot of attention in recent studies of machine performances. To alleviate the strong assumption in learning. These methods project data into high-dimensional the seeking of the approximation bounds, we take a more or even infinite-dimensional spaces via kernel mapping general assumption that the design matrix K ensuring the restricted functions. Despite the strong generalization ability induced isometric property (Koltchinskii 2011). In particular, by kernel methods, they usually suffer from the high computation the new assumption obeys the restricted eigenvalue condition complexity of calculating the kernel matrix (also (Koltchinskii 2011; Bickel, Ritov, and Tsybakov 2009), called Gram matrix). Although low-rank decomposition which has been shown to be more general than several techniques(e.g., Cholesky Decomposition (Fine and Scheinberg other similar assumptions used in sparsity literature (Candes 2002; Bach and Jordan 2005)), and truncating methods(e.g., and Tao 2007; Donoho, Elad, and Temlyakov 2006; Kernel Tapering (Shen, Xu, and Allebach 2014; Zhang and Huang 2008). Based on the restricted eigenvalue Furrer, Genton, and Nychka 2006)) can accelerate the calculation condition, we have provided error bounds for kernel approximation of the kernel matrix, they still need to compute the and recovery rate in sparse kernel regression.

Naive Bayes (NB) is well-known to be a simple but effective classifier, especially when combined with feature selection. Unfortunately, feature selection methods are often greedy and thus cannot guarantee an optimal feature set is selected. An alternative to feature selection is to use Bayesian model averaging (BMA), which computes a weighted average over multiple predictors; when the different predictor models correspond to different feature sets, BMA has the advantage over feature selection that its predictions tend to have lower variance on average in comparison to any single model. In this paper, we show for the first time that it is possible to exactly evaluate BMA over the exponentially-sized powerset of NB feature models in linear-time in the number of features; this yields an algorithm about as expensive to train as a single NB model with all features, but yet provably converges to the globally optimal feature subset in the asymptotic limit of data. We evaluate this novel BMA-NB classifier on a range of datasets showing that it never underperforms NB (as expected) and sometimes offers performance competitive (or superior) to classifiers such as SVMs and logistic regression while taking a fraction of the time to train.