郝新丽

(Cloud Group)

报告题目：BRITS: Bidirectional Recurrent Imputation for Time Series

报告摘要：

Time series are widely used as signals in many classification/regression tasks. It is ubiquitous that time series contains many missing values. Given multiple correlated time series data, how to fill in missing values and to predict their class labels? Existing imputation methods often impose strong assumptions of the underlying data generating process, such as linear dynamics in the state space. In this paper, we propose BRITS, a novel method based on recurrent neural networks for missing value imputation in time series data. Our proposed method directly learns the missing values in a bidirectional recurrent dynamical system, without any specific assumption. The imputed values are treated as variables of RNN graph and can be effectively updated during the backpropagation.BRITS has three advantages: (a) it can handle multiple correlated missing values in time series; (b) it generalizes to time series with nonlinear dynamics underlying; (c) it provides a data-driven imputation procedure and applies to general settings with missing data.

2020.7.3  腾讯会议

郝新丽

(Cloud Group)

报告题目：Robust and Rapid Clustering of KPIs for Large-Scale Anomaly Detection

报告摘要：For large Internet companies, it is very important to monitor a large number of KPIs (Key Performance Indicators) and detect anomalies to ensure the service quality and reliability. However, large-scale anomaly detection on millions of KPIs is very challenging due to the large overhead of model selection, parameter tuning, model training, or labeling. In this paper we argue that KPI clustering can help: we can cluster millions of KPIs into a small number of clusters and then select and train model on a per-cluster basis. However, KPI clustering faces new challenges that are not present in classic time series clustering: KPIs are typically much longer than other time series, and noises, anomalies, phase shifts and amplitude differences often change the shape of KPIs and mislead the clustering algorithm.

2020.10.13  FL1, Wing Building for Science Complex

但唐朋

(Cloud Group)

报告题目：Spatial Temporal Trajectory Similarity Join

报告摘要：Existing works only focus on spatial dimension without the consideration of combining spatial and temporal dimensions together when processing trajectory similarity join queries, to address this problem, this paper proposes a novel two-level grid index which takes both spatial and temporal information into account when processing spatial-temporal trajectory similarity join. A new similarity function MOGS is developed to measure the similarity in an efficient manner when our candidate trajectories have high coverage rate CR. Extensive experiments are conducted to verify the efficiency of our solution.

2020.10.20  FL1, Wing Building for Science Complex

但唐朋

(Cloud Group)

报告题目：Searching Activity Trajectories by Exemplar

报告摘要：The rapid explosion of urban cities has modernized the residents’ lives and generated a large amount of data (e.g., human mobility data, traffic data, and geographical data), especially the activity trajectory data that contains spatial and temporal as well as activity information. With these data, urban computing enables to provide better services such as location-based applications for smart cities. Recently, a novel exemplar query paradigm becomes popular that considers a user query as an example of the data of interest, which plays an important role in dealing with the information deluge. In this article, we propose a novel query, called searching activity trajectory by exemplar, where, given an exemplar trajectory τq, the goal is to find the top-k trajectories with the smallest distances to τq. We first introduce an inverted-index-based algorithm (ILA) using threshold ranking strategy. To further improve the efficiency, we propose a gridtree threshold approach (GTA) to quickly locate candidates and prune unnecessary trajectories. In addition, we extend GTA to support parallel processing. Finally, extensive experiments verify the high efficiency and scalability of the proposed algorithms.

2020.11.3  FL1, Wing Building for Science Complex

彭迎涛

(Web Group)

报告题目：RippleNet: Propagating User Preferences on the Knowledge Graph for Recommender Systems

报告摘要：To address the sparsity and cold start problem of collaborative filtering, researchers usually make use of side information, such as social networks or item attributes, to improve recommendation performance. This paper considers the knowledge graph as the source of side information. To address the limitations of existing embeddingbased and path-based methods for knowledge-graph-aware recommendation, we propose RippleNet, an end-to-end framework that naturally incorporates the knowledge graph into recommender systems. Similar to actual ripples propagating on the water, RippleNet stimulates the propagation of user preferences over the set of knowledge entities by automatically and iteratively extending a user’s potential interests along links in the knowledge graph. The multiple "ripples" activated by a user’s historically clicked items are thus superposed to form the preference distribution of the user with respect to a candidate item, which could be used for predicting the final clicking probability. Through extensive experiments on real-world datasets, we demonstrate that RippleNet achieves substantial gains in a variety of scenarios, including movie, book and news recommendation, over several state-of-the-art baselines.

2020.11.10  FL1, Wing Building for Science Complex

郝新丽

(Cloud Group)

报告题目：Opprentice: Towards Practical and Automatic Anomaly Detection Through Machine Learning

报告摘要：Closely monitoring service performance and detecting anomalies are critical for Internet-based services. However, even though dozens of anomaly detectors have been proposed over the years, deploying them to a given service remains a great challenge, requiring manually and iteratively tuning detector parameters and thresholds. This paper tackles this challenge through a novel approach based on supervised machine learning. With our proposed system, Opprentice (Operators’ apprentice), operators’ only manual work is to periodically label the anomalies in the performance data with a convenient tool. Multiple existing detectors are applied to the performance data in parallel to extract anomaly features. Then the features and the labels are used to train a random forest classifier to automatically select the appropriate detector-parameter combinations and the thresholds. For three different service KPIs in a top global search engine, Opprentice can automatically satisfy or approximate a reasonable accuracy preference (recall ≥ 0.66 and precision ≥ 0.66). More importantly, Opprentice allows operators to label data in only tens of minutes, while operators traditionally have to spend more than ten days selecting and tuning detectors, which may still turn out not to work in the end.

马超红

(Cloud Group)

报告题目：Learning Multi-dimensional Indexes

报告摘要：Scanning and filtering over multi-dimensional tables are key operations in modern analytical database engines. To optimize the performance of these operations, databases often create clustered indexes over a single dimension or multidimensional indexes such as R-Trees, or use complex sort orders (e.g., Z-ordering). However, these schemes are often hard to tune and their performance is inconsistent across different datasets and queries. This paper introduce Flood, a multi-dimensional in-memory read-optimized index that automatically adapts itself to a particular dataset and workload by jointly optimizing the index structure and data storage layout. Flood a new multi-dimensional primary index that is jointly optimized using both the underlying data and query workloads.

但唐朋

(Cloud Group)

报告题目：Neural circuit policies enabling auditable autonomy

报告摘要：A central goal of artificial intelligence in high-stakes decision-making applications is to design a single algorithm that simultaneously expresses generalizability by learning coherent representations of their world and interpretable explanations of its dynamics. Here, we combine brain-inspired neural computation principles and scalable deep learning architectures to design compact neural controllers for task-specific compartments of a full-stack autonomous vehicle control system. We discover that a single algorithm with 19 control neurons, connecting 32 encapsulated input features to outputs by 253 synapses, learns to map high-dimensional inputs into steering commands. This system shows superior generalizability, interpretability and robustness compared with orders-of-magnitude larger black-box learning systems. The obtained neural agents enable high-fidelity autonomy for task-specific parts of a complex autonomous system.

2020.11.17  FL1, Wing Building for Science Complex

郝新丽

(Cloud Group)

报告题目：Active Model Selection for Positive Unlabeled Time Series Classification

报告摘要：Positive unlabeled time series classification (PUTSC) refers to classifying time series with a set P L of positive labeled examples and a set U of unlabeled ones. Model selection for PUTSC is a largely untouched topic. In this paper, we look into PUTSC model selection, which as far as we know is the first systematic study in this topic. Focusing on the widely adopted self-training one-nearest-neighbor (ST-1NN) paradigm, we propose a model selection framework based on active learning (AL). We present the novel concepts of self-training label propagation, pseudo label calibration principles and ultimately influence to fully exploit the mechanism of ST-1NN. Based on them, we develop an effective model performance evaluation strategy and three AL sampling strategies. Experiments on over 120 datasets and a case study in arrhythmia detection show that our methods can yield top performance in interactive environments, and can achieve near optimal results by querying very limited numbers of labels from the AL oracle.

马超红

(Cloud Group)

报告题目：DBOS: A Database-oriented operating system

报告摘要：Current operating systems are complex systems that were designed before today's computing environments. This makes it difficult for them to meet the scalability, heterogeneity, availability, and security challenges in current cloud and parallel computing environments. To address these problems, this paper propose a radically new OS design based on data-centric architecture: all operating system state should be represented uniformly as database tables, and operations on this state should be made via queries from otherwise stateless tasks. This design makes it easy to scale and evolve the OS without whole-system refactoring, inspect and debug system state, upgrade components without downtime, manage decisions using machine learning, and implement sophisticated security features. Everything is table, every request is query.

2020.11.24  FL1, Wing Building for Science Complex

彭迎涛

(Web Group)

报告题目：Knowledge Graph Convolutional Networks for Recommender Systems

报告摘要：To alleviate sparsity and cold start problem of collaborative filtering based recommender systems, researchers and engineers usually collect attributes of users and items, and design delicate algorithms to exploit these additional information. In general, the attributes are not isolated but connected with each other, which forms a knowledge graph (KG). In this paper, we propose Knowledge Graph Convolutional Networks (KGCN), an end-to-end framework that captures inter-item relatedness effectively by mining their associated attributes on the KG. To automatically discover both high-order structure information and semantic information of the KG, we sample from the neighbors for each entity in the KG as their receptive field, then combine neighborhood information with bias when calculating the representation of a given entity. The receptive field can be extended to multiple hops away to model high-order proximity information and capture users’ potential long-distance interests. Moreover, we implement the proposed KGCN in a minibatch fashion, which enables our model to operate on large datasets and KGs. We apply the proposed model to three datasets about movie, book, and music recommendation, and experiment results demonstrate that our approach outperforms strong recommender baselines

2020.12.1  FL1, Wing Building for Science Complex

彭迎涛

(Web Group)

报告题目：KGAT: Knowledge Graph Attention Network for Recommendation      

报告摘要：To provide more accurate, diverse, and explainable recommendation, it is compulsory to go beyond modeling user-item interactions and take side information into account. Traditional methods like factorization machine (FM) cast it as a supervised learning problem, which assumes each interaction as an independent instance with side information encoded. Due to the overlook of the relations among instances or items (e.g., the director of a movie is also an actor of another movie), these methods are insufficient to distill the collaborative signal from the collective behaviors of users. In this work, we investigate the utility of knowledge graph (KG), which breaks down the independent interaction assumption by linking items with their attributes. We argue that in such a hybrid structure of KG and user-item graph, high-order relations — which connect two items with one or multiple linked attributes — are an essential factor for successful recommendation. We propose a new method named Knowledge Graph Attention Network (KGAT) which explicitly models the high-order connectivities in KG in an end-to-end fashion. It recursively propagates the embeddings from a node’s neighbors (which can be users, items, or attributes) to refine the node’s embedding, and employs an attention mechanism to discriminate the importance of the neighbors. Our KGAT is conceptually advantageous to existing KG-based recommendation methods, which either exploit high-order relations by extracting paths or implicitly modeling them with regularization. Empirical results on three public benchmarks show that KGAT significantly outperforms state-of-the-art methods like Neural FM and RippleNet. Further studies verify the efficacy of embedding propagation for high-order relation modeling and the interpretability benefits brought by the attention mechanism.  

2020.12.08  FL1, Wing Building for Science Complex

但唐朋

(Cloud Group)

报告题目：Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learning  

报告摘要：Increasing the scale of reinforcement learning experiments has allowed researchers to achieve unprecedented results in both training sophisticated agents for video games, and in sim-to-real transfer for robotics. Typically such experiments rely on large distributed systems and require expensive hardware setups, limiting wider access to this exciting area of research. In this work we aim to solve this problem by optimizing the efficiency and resource utilization of reinforcement learning algorithms instead of relying on distributed computation. We present the "Sample Factory", a high-throughput training system optimized for a single-machine setting. Our architecture combines a highly efficient, asynchronous, GPU-based sampler with off-policy correction techniques, allowing us to achieve throughput higher than 10^5 environment frames/second on non-trivial control problems in 3D without sacrificing sample efficiency. We extend Sample Factory to support self-play and population-based training and apply these techniques to train highly capable agents for a multiplayer first-person shooter game.

2021.01.05  FL1, Wing Building for Science Complex

彭迎涛

(Web Group)

报告题目：IRGAN: A Minimax Game for Unifying Generative and Discriminative Information Retrieval Models

报告摘要：

This paper provides a unified account of two schools of thinking in information retrieval modelling: the generative retrieval focusing on predicting relevant documents given a query, and the discriminative retrieval focusing on predicting relevancy given a query-document pair. We propose a game theoretical minimax game to iteratively optimise both models. On one hand, the discriminative model, aiming to mine signals from labelled and unlabelled data, provides guidance to train the generative model towards fitting the underlying relevance distribution over documents given the query. On the other hand, the generative model, acting as an attacker to the current discriminative model, generates difficult examples for the discriminative model in an adversarial way by minimising its discrimination objective. With the competition between these two models, we show that the unified framework takes advantage of both schools of thinking: (i) the generative model learns to fit the relevance distribution over documents via the signals from the discriminative model, and (ii) the discriminative model is able to exploit the unlabelled data selected by the generative model to achieve a better estimation for document ranking.