Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Data Mining - Concepts and Techniques.

/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

The PASCAL Visual Object Classes Challenge

Transparent conducting ZnO:Al films deposited on organic substrates deposited by r.f. magnetron-sputtering

Session-based recommendation is the task of predicting the next item to recommend when the only available information consists of anonymous behavior sequences. Previous methods for session-based recommendation focus mostly on the current session, ignoring collaborative information in so-called neighborhood sessions, sessions that have been generated previously by other users and reflect similar user intents as the current session. We hypothesize that the collaborative information contained in such neighborhood sessions may help to improve recommendation performance for the current session. We propose a Collaborative Session-based Recommendation Machine (CSRM), a novel hybrid framework to apply collaborative neighborhood information to session-based recommendations. CSRM consists of two parallel modules: an Inner Memory Encoder (IME) and an Outer Memory Encoder (OME). The IME models a user's own information in the current session with the help of Recurrent Neural Networks (RNNs) and an attention mechanism. The OME exploits collaborative information to better predict the intent of current sessions by investigating neighborhood sessions. Then, a fusion gating mechanism is used to selectively combine information from the IME and OME so as to obtain the final representation of the current session. Finally, CSRM obtains a recommendation score for each candidate item by computing a bilinear match with the final representation. Experimental results on three public datasets demonstrate the effectiveness of CSRM compared to state-of-the-art session-based recommender systems. Our analysis of CSRM's recommendation process shows when and how collaborative neighborhood information and the fusion gating mechanism positively impact the performance of session-based recommendations.

A Collaborative Session-based Recommendation Approach with Parallel Memory Modules

Ranking nodes by their spreading ability in complex networks is a fundamental problem which relates to wide applications. Local metric like degree centrality is simple but less effective. Global metrics such as betweenness and closeness centrality perform well in ranking nodes, but are of high computational complexity. Recently, to rank nodes effectively and efficiently, a semi-local centrality measure has been proposed as a tradeoff between local and global metrics. However, in semi-local centrality, only the number of the nearest and the next nearest neighbors of a node is taken into account, while the topological connections among the neighbors are neglected. In this paper, we propose a local structural centrality measure which considers both the number and the topological connections of the neighbors of a node. To evaluate the performance of our method, we use the Susceptible–Infected–Recovered (SIR) model to simulate the epidemic spreading process on both artificial and real networks. By measuring the rank correlation between the ranked list generated by simulation results and the ones generated by centrality measures, we show that our method can rank the spreading ability of nodes more accurately than centrality measures such as degree, k -shell, betweenness, closeness and local centrality. Further, we show that our method can better distinguish the spreading ability of nodes.

Ranking the spreading ability of nodes in complex networks based on local structure

Popularity prediction on microblogging platforms aims to predict the future popularity of a message based on its retweeting dynamics in the early stages. Existing works mainly focus on exploring effective features for prediction, while ignoring the underlying arrival process of retweets. Also, the effect of user activity variation on the retweeting dynamics in the early stages has been neglected. In this paper, we propose an extended reinforced Poisson process model with time mapping process to model the retweeting dynamics and predict the future popularity. The proposed model explicitly characterizes the process through which a message gain its retweets, by capturing a power-law temporal relaxation function corresponding to the aging in the ability of the message to attract new retweets and an exponential reinforcement mechanism characterizing the "richer-get-richer" phenomenon. Further, we introduce the notation of weibo time and integrate a time mapping process into the proposed model to eliminate the effect of user activity variation. Extensive experiments on two Weibo datasets, with 10K and 18K messages respectively, well demonstrate the effectiveness of our proposed model in popularity prediction.

Modeling and Predicting Retweeting Dynamics on Microblogging Platforms

Sequential Recommendation (SR) is the task of recommending the next item based on a sequence of recorded user behaviors. We study SR in a particularly challenging context, in which multiple individual users share a single account (shared-account) and in which user behaviors are available in multiple domains (cross-domain). These characteristics bring new challenges on top of those of the traditional SR task. On the one hand, we need to identify different user behaviors under the same account in order to recommend the right item to the right user at the right time. On the other hand, we need to discriminate the behaviors from one domain that might be helpful to improve recommendations in the other domains. We formulate the Shared-account Cross-domain Sequential Recommendation (SCSR) task as a parallel sequential recommendation problem. We propose a Parallel Information-sharing Network (π-Net) to simultaneously generate recommendations for two domains where user behaviors on two domains are synchronously shared at each timestamp. π-Net contains two core units: a shared account filter unit (SFU) and a cross-domain transfer unit (CTU). The SFU is used to address the challenge raised by shared accounts; it learns user-specific representations, and uses a gating mechanism to filter out information of some users that might be useful for another domain. The CTU is used to address the challenge raised by the cross-domain setting; it adaptively combines the information from the SFU at each timestamp and then transfers it to another domain. After that, π-Net estimates recommendation scores for each item in two domains by integrating information from both domains. To assess the effectiveness of π-Net, we construct a new dataset HVIDEO from real-world smart TV watching logs. To the best of our knowledge, this is the first dataset with both shared-account and cross-domain characteristics. We release HVIDEO to facilitate future research. Our experimental results demonstrate that π-Net outperforms state-of-the-art baselines in terms of MRR and Recall.

Jun Ma

Papers

Transparent conducting ZnO:Al films deposited on organic substrates deposited by r.f. magnetron-sputtering

A Collaborative Session-based Recommendation Approach with Parallel Memory Modules

Ranking the spreading ability of nodes in complex networks based on local structure

Modeling and Predicting Retweeting Dynamics on Microblogging Platforms

π-Net: A Parallel Information-sharing Network for Shared-account Cross-domain Sequential Recommendations