Temporal action localization, which aims at recognizing the location and the category of action instances in videos, has long been researched. Existing methods divide each video into multiple action units (i.e., proposals in two-stage methods and segments in one-stage methods) and then perform recognition/regression on each of them individually without explicitly exploiting their relations, which, however, play an important role in action localization. In this paper, we propose a general graph convolutional module (GCM) that can be easily plugged into existing action localization methods, including two-stage and one-stage paradigms. Specifically, we first construct a graph, where each action unit is represented as a node and their relations as edges. We use two types of relations, one for capturing the temporal connections, and the other one for characterizing the semantic relationship. Then, we apply graph convolutional networks (GCNs) on the graph to model the relations and learn more informative representations for action localization. Experimental results show that GCM consistently improves the performance of both two-stage action localization methods (e.g., CBR and R-C3D) and one-stage methods (e.g., D-SSAD), verifying the generality and effectiveness of GCM. Moreover, with the aid of GCM, our approach significantly outperforms the state-of-the-art on THUMOS14 and ActivityNet.

Portfolio Selection is an important real-world financial task and has attracted extensive attention in artificial intelligence communities. This task, however, has two main difficulties: (i) the non-stationary price series and complex asset correlations make the learning of feature representation very hard; (ii) the practicality principle in financial markets requires controlling both transaction and risk costs. Most existing methods adopt handcraft features and consider no constraints for the costs, which may make them perform unsatisfactorily and fail to control both costs in practice. In this paper, we propose a cost-sensitive portfolio selection method with deep reinforcement learning. Specifically, a novel two-stream portfolio policy network is devised to extract both price series patterns and asset correlations, while a new cost-sensitive reward function is developed to maximize the accumulated return and constrain both costs via reinforcement learning. We theoretically analyze the near-optimality of the proposed reward, which shows that the growth rate of the policy regarding this reward can approach the theoretical optimum. We also empirically evaluate the proposed method on real-world datasets. Promising results demonstrate the effectiveness and superiority of the proposed method in terms of profitability, cost-sensitivity and representation abilities.

Cost-Sensitive Online Classification has drawn extensive attention in recent years, where the main approach is to directly online optimize two well-known cost-sensitive metrics: (i) weighted sum of sensitivity and specificity; (ii) weighted misclassification cost. However, previous existing methods only considered first-order information of data stream. It is insufficient in practice, since many recent studies have proved that incorporating second-order information enhances the prediction performance of classification models. Thus, we propose a family of cost-sensitive online classification algorithms with adaptive regularization in this paper. We theoretically analyze the proposed algorithms and empirically validate their effectiveness and properties in extensive experiments. Then, for better trade off between the performance and efficiency, we further introduce the sketching technique into our algorithms, which significantly accelerates the computational speed with quite slight performance loss. Finally, we apply our algorithms to tackle several online anomaly detection tasks from real world. Promising results prove that the proposed algorithms are effective and efficient in solving cost-sensitive online classification problems in various real-world domains.

The probability prediction of multivariate time series is a notoriously challenging but practical task. On the one hand, the challenge is how to effectively capture the cross-series correlations between interacting time series, to achieve accurate distribution modeling. On the other hand, we should consider how to capture the contextual information within time series more accurately to model multivariate temporal dynamics of time series. In this work, we proposed a novel non-autoregressive deep learning model, called Multi-scale Attention Normalizing Flow(MANF), where we combine multi-scale attention with relative position information and the multivariate data distribution is represented by the conditioned normalizing flow. Additionally, compared with autoregressive modeling methods, our model avoids the influence of cumulative error and does not increase the time complexity. Extensive experiments demonstrate that our model achieves state-of-the-art performance on many popular multivariate datasets.

The main target of retrosynthesis is to recursively decompose desired molecules into available building blocks. Existing template-based retrosynthesis methods follow a template selection stereotype and suffer from limited training templates, which prevents them from discovering novel reactions. To overcome this limitation, we propose an innovative retrosynthesis prediction framework that can compose novel templates beyond training templates. As far as we know, this is the first method that uses machine learning to compose reaction templates for retrosynthesis prediction. Besides, we propose an effective reactant candidate scoring model that can capture atom-level transformations, which helps our method outperform previous methods on the USPTO-50K dataset. Experimental results show that our method can produce novel templates for 15 USPTO-50K test reactions that are not covered by training templates. We have released our source implementation.

The goal of online active learning is to learn predictive models from a sequence of unlabeled data given limited label query budget. Unlike conventional online learning tasks, online active learning is considerably more challenging since the online learner not only needs to decide how to update the predictive models effectively whenever a labeled training sample is available, but also needs to determine when is appropriate to query the label of an incoming unlabeled instance given limited label query budget. Most existing approaches for online active learning are often based on the family of first-order online learning algorithms, which are simple and efficient but falls short in slow convergence and sub-optimal solution in exploiting the labeled training data. To conquer these limitations, this paper presents a novel framework of Second-order Online Active Learning (SOAL). To further make SOAL practical for real-world applications, especially for class-imbalanced online classification tasks (e.g., malicious web detection), we extend the SOAL framework by proposing the Cost-sensitive Second-order Online Active Learning algorithm named “SOAL$_{CS}$”. We conducted both theoretical analysis and empirical studies, in which the promising empirical results validate the efficacy and scalability of the proposed online active learning algorithms towards large-scale online learning tasks.