Deep supervised hashing (DSH), which combines binary learning and convolutional neural network, has attracted considerable research interests and achieved promising performance for highly efficient image retrieval. In this paper, we show that the widely used loss functions, pair-wise loss and triplet loss, suffer from the trivial solution problem and usually lead to highly correlated bits in practice, limiting the performance of DSH. One important reason is that it is difficult to incorporate proper constraints into the loss functions under the mini-batch based optimization algorithm. To tackle these problems, we propose to adopt ensemble learning strategy for deep model training. We found out that this simple strategy is capable of effectively decorrelating different bits, making the hashcodes more informative. Moreover, it is very easy to parallelize the training and support incremental model learning, which are very useful for real-world applications but usually ignored by existing DSH approaches. Experiments on benchmarks demonstrate the proposed ensemble based DSH can improve the performance of DSH approaches significant.

Search task success rate is a crucial metric based on the search experience of users to measure the performance of search systems. Modeling search action sequence would help to capture the latent search patterns of users in successful and unsuccessful search tasks. Existing approaches use aggregated features to describe the user behavior in search action sequences, which depend on heuristic hand-crafted feature design and ignore a lot of information inherent in the user behavior. In this paper, we employ Long Short-Term Memory (LSTM) that performs end-to-end fine-tuning during the training to learn search action sequence representation for search task success evaluation. Concretely, we normalize the search action sequences by introducing a dummy idle action, which guarantees that the time intervals between contiguous actions are fixed. Simultaneously, we propose a novel data augmentation strategy to increase the pattern variations on search action sequence data to improve the generalization ability of LSTM. We evaluate the proposed approach on open datasets with two different definitions of search task success. The experimental results show that the proposed approach achieves significant performance improvement compared with several excellent search task success evaluation approaches.

Molecular dynamics (MD) is a powerful computational method for simulating molecular behavior. Deep neural networks provide a novel method of generating MD data efficiently, but there is no architecture that mitigates the well-known exposure bias accumulated by multi-step generations. In this paper, we propose a multi-step time series generator using a deep neural network based on Wasserstein generative adversarial nets. Instead of sparse real data, our model evolves a latent variable z that is densely distributed in a low-dimensional space. This novel framework successfully mitigates the exposure bias. Moreover, our model can evolve part of the system (Feature extraction) with any time step (Step skip), which accelerates the efficient generation of MD data. The applicability of this model is evaluated through three different systems: harmonic oscillator, bulk water, and polymer melts. The experimental results demonstrate that our model can generate time series of the MD data with sufficient accuracy to calculate the physical and important dynamical statistics.

How does the machine classify styles in art? And how does it relate to art historians' methods for analyzing style? Several studies showed the ability of the machine to learn and predict styles, such as Renaissance, Baroque, Impressionism, etc., from images of paintings. This implies that the machine can learn an internal representation encoding discriminative features through its visual analysis. However, such a representation is not necessarily interpretable. We conducted a comprehensive study of several of the state-of-the-art convolutional neural networks applied to the task of style classification on 67K images of paintings, and analyzed the learned representation through correlation analysis with concepts derived from art history. Surprisingly, the networks could place the works of art in a smooth temporal arrangement mainly based on learning style labels, without any a priori knowledge of time of creation, the historical time and context of styles, or relations between styles. The learned representations showed that there are a few underlying factors that explain the visual variations of style in art. Some of these factors were found to correlate with style patterns suggested by Heinrich Wölfflin (1846-1945). The learned representations also consistently highlighted certain artists as the extreme distinctive representative of their styles, which quantitatively confirms art historian observations.

Deep neural networks have shown promise in collaborative filtering (CF). However, existing neural approaches are either user-based or item-based, which cannot leverage all the underlying information explicitly. We propose CF-UIcA, a neural co-autoregressive model for CF tasks, which exploits the structural correlation in the domains of both users and items. The co-autoregression allows extra desired properties to be incorporated for different tasks. Furthermore, we develop an efficient stochastic learning algorithm to handle large scale datasets. We evaluate CF-UIcA on two popular benchmarks: MovieLens 1M and Netflix, and achieve state-of-the-art performance in both rating prediction and top-N recommendation tasks, which demonstrates the effectiveness of CF-UIcA.

Learning low-dimensional representations of networks has proved effective in a variety of tasks such as node classification, link prediction and network visualization. Existing methods can effectively encode different structural properties into the representations, such as neighborhood connectivity patterns, global structural role similarities and other high-order proximities. However, except for objectives to capture network structural properties, most of them suffer from lack of additional constraints for enhancing the robustness of representations. In this paper, we aim to exploit the strengths of generative adversarial networks in capturing latent features, and investigate its contribution in learning stable and robust graph representations. Specifically, we propose an Adversarial Network Embedding (ANE) framework, which leverages the adversarial learning principle to regularize the representation learning. It consists of two components, i.e., a structure preserving component and an adversarial learning component. The former component aims to capture network structural properties, while the latter contributes to learning robust representations by matching the posterior distribution of the latent representations to given priors. As shown by the empirical results, our method is competitive with or superior to state-of-the-art approaches on benchmark network embedding tasks.

We propose an end-to-end approach for modeling polyphonic music with a novel graphical representation, based on music theory, in a deep neural network. Despite the success of deep learning in various applications, it remains a challenge to incorporate existing domain knowledge in a network without affecting its training routines. In this paper we present a novel approach for predictive music modeling and music generation that incorporates domain knowledge in its representation. In this work, music is transformed into a 2D representation, inspired by tonnetz from music theory, which graphically encodes musical relationships between pitches. This representation is incorporated in a deep network structure consisting of multilayered convolutional neural networks (CNN, for learning an efficient abstract encoding of the representation) and recurrent neural networks with long short-term memory cells (LSTM, for capturing temporal dependencies in music sequences). We empirically evaluate the nature and the effectiveness of the network by using a dataset of classical music from various composers. We investigate the effect of parameters including the number of convolution feature maps, pooling strategies, and three configurations of the network: LSTM without CNN, LSTM with CNN (pre-trained vs. not pre-trained). Visualizations of the feature maps and filters in the CNN are explored, and a comparison is made between the proposed tonnetz-inspired representation and pianoroll, a commonly used representation of music in computational systems. Experimental results show that the tonnetz representation produces musical sequences that are more tonally stable and contain more repeated patterns than sequences generated by pianoroll-based models, a finding that is directly useful for tackling current challenges in music and AI such as smart music generation.

Urban air pollution has attracted much attention these years for its adverse impacts on human health. While monitoring stations have been established to collect pollutant statistics, the number of stations is very limited due to the high cost. Thus, inferring fine-grained urban air quality information is becoming an essential issue for both government and people. In this paper, we propose a generic neural approach, named ADAIN, for urban air quality inference. We leverage both the information from monitoring stations and urban data that are closely related to air quality, including POIs, road networks and meteorology. ADAIN combines feedforward and recurrent neural networks for modeling static and sequential features as well as capturing deep feature interactions effectively. A novel attempt of ADAIN is an attention-based pooling layer that automatically learns the weights of features from different monitoring stations, to boost the performance. We conduct experiments on a real-world air quality dataset and our approach achieves the highest performance compared with various state-of-the-art solutions.

With the growing volume of publications in the Computer Science (CS) discipline, tracking the research evolution and predicting the future research trending topics are of great importance for researchers to keep up with the rapid progress of research. Within a research area, there are many top conferences that publish the latest research results. These conferences mutually influence each other and jointly promote the development of the research area. To predict the trending topics of mutually influenced conferences, we propose a correlated neural influence model, which has the ability to capture the sequential properties of research evolution in each individual conference and discover the dependencies among different conferences simultaneously. The experiments conducted on a scientific dataset including conferences in artificial intelligence and data mining show that our model consistently outperforms the other state-of-the-art methods. We also demonstrate the interpretability and predictability of the proposed model by providing its answers to two questions of concern, i.e., what the next rising trending topics are and for each conference who the most influential peer is.

We train and validate a semi-supervised, multi-task LSTM on 57,675 person-weeks of data from off-the-shelf wearable heart rate sensors, showing high accuracy at detecting multiple medical conditions, including diabetes (0.8451), high cholesterol (0.7441), high blood pressure (0.8086), and sleep apnea (0.8298). We compare two semi-supervised training methods, semi-supervised sequence learning and heuristic pretraining, and show they outperform hand-engineered biomarkers from the medical literature. We believe our work suggests a new approach to patient risk stratification based on cardiovascular risk scores derived from popular wearables such as Fitbit, Apple Watch, or Android Wear.