Cascade Ranking for Operational E-commerce Search Liu, Shichen; Xiao, Fei; Ou, Wenwu ...
Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,
08/2017
Conference Proceeding
Open access
In the 'Big Data' era, many real-world applications like search involve the ranking problem for a large number of items. It is important to obtain effective ranking results and at the same time ...obtain the results efficiently in a timely manner for providing good user experience and saving computational costs. Valuable prior research has been conducted for learning to efficiently rank like the cascade ranking (learning) model, which uses a sequence of ranking functions to progressively filter some items and rank the remaining items. However, most existing research of learning to efficiently rank in search is studied in a relatively small computing environments with simulated user queries.
This paper presents novel research and thorough study of designing and deploying a Cascade model in a Large-scale Operational E-commerce Search application (CLOES), which deals with hundreds of millions of user queries per day with hundreds of servers. The challenge of the real-world application provides new insights for research: 1). Real-world search applications often involve multiple factors of preferences or constraints with respect to user experience and computational costs such as search accuracy, search latency, size of search results and total CPU cost, while most existing search solutions only address one or two factors; 2). Effectiveness of e-commerce search involves multiple types of user behaviors such as click and purchase, while most existing cascade ranking in search only models the click behavior. Based on these observations, a novel cascade ranking model is designed and deployed in an operational e-commerce search application. An extensive set of experiments demonstrate the advantage of the proposed work to address multiple factors of effectiveness, efficiency and user experience in the real-world application.
Perceive Your Users in Depth Ni, Yabo; Ou, Dan; Liu, Shichen ...
Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining,
07/2018
Conference Proceeding
Tasks such as search and recommendation have become increasingly important for E-commerce to deal with the information overload problem. To meet the diverse needs of different users, personalization ...plays an important role. In many large portals such as Taobao and Amazon, there are a bunch of different types of search and recommendation tasks operating simultaneously for personalization. However, most of current techniques address each task separately. This is suboptimal as no information about users shared across different tasks.
In this work, we propose to learn universal user representations across multiple tasks for more effective personalization. In particular, user behavior sequences (e.g., click, bookmark or purchase of products) are modeled by LSTM and attention mechanism by integrating all the corresponding content, behavior and temporal information. User representations are shared and learned in an end-to-end setting across multiple tasks. Benefiting from better information utilization of multiple tasks, the user representations are more effective to reflect their interests and are more general to be transferred to new tasks. We refer this work as Deep User Perception Network (DUPN) and conduct an extensive set of offline and online experiments. Across all tested five different tasks, our DUPN consistently achieves better results by giving more effective user representations. Moreover, we deploy DUPN in large scale operational tasks in Taobao. Detailed implementations, e.g., incremental model updating, are also provided to address the practical issues for the real world applications.
To evaluate the significance of temporary block of gastroduodenal artery in hepatic artery chemotherapy.
Forty patients were randomized into two groups with 20 in each. In the trial group, when the ...catheter was introduced into the gastroduodenal artery, pituitrin was infused slowly (2 U/min) till the gastroduodenal artery became blocked, then the catheter was pulled back to the common hepatic artery to start chemotherapy. In the control group, saline (10 ml) was infused slowly (4 ml/min) instead of pituitrin.
In the trial group, all patients had temporary increase of blood pressure ranging from 20 to 50 mm Hg, which was tolerated with most recovered in 20 to 30 minutes. Two patients had pain in the upper abdomen and others only had slight gastrointestinal discomfort. In the control group, epigastric upset or pain during operation was present in 5 patients. In 9 patients, upper abdominal pain after the operation was present which was serious in two. One of these two patients was confirmed as having gastric ant
Buried hill reservoir is indispensable part of karst reservoirs in northern Tarim basin, China, which is resulted from the long-term exposure of Cambrian carbonate rock due to meteoric freshwater. ...Four types of zones could be divided in the buried hill of Upper Cambrian Xiaqiulitage Formation in Yingmai-Yaha structure, including surface karst zone, vertical vadose zone, horizontal underflow zone and deep slow flow zone from up to bottom. Firstly, the conventional and image log response characteristics of different types of vertical karst zonation were summarized through cores, thin section and well log-calibration. The surface karst zone is close to the weathering crust, showing breccia and residual characteristics, and is recognized as bright spots appearances on the image logs. Due to the vertical movement of atmospheric fresh water, a large number of high-angle fractures and dissolution caves are developed in the vertical vadose zone, and the vertical vadose zones are characterized by high-angle sine curve appearances on the image logs. The horizontal under flow zone has a horizontally connected caves or vugs due to the horizontal movement of groundwater, and the image logs are recognized as dark bands. The conventional logs also show the decrease of resistivity and the increase of gamma ray (GR). The deep slow flow zone is less affected by karstification, and there are fewer dissolution caves and fractures, but there is also a certain weak dissolution and may contain small vugs. The image logs show isolated dark spots. The Vertical zonation of buried hill can be divided using well logs by calibrating cores, and thin sections. Finally, the core analysis, and the oil test data are used to determine the zoning of the high-quality reservoir. The results show that the four zones of the buried hills have obvious well log response characteristics. The karst zoning can be well divided by the image logs and conventional well logs when calibrated with cores. The data of core analysis and oil testing show that although the reservoir quality in the horizontal underflow zone is the best, the study area is a bottom-water reservoir, and the reservoirs in the horizontal underflow zone are mainly water-producing. On the contrary, in the vertical vadose zone, a large number of high-angle fractures communicated with small pores, which not only improved the quality of the reservoir, but also the oil test data proves that the reservoir is mainly producing oil and gas. Therefore, compared with the horizontal underflow zone and deep slow flow zone, the vertical vadose zone is favorable to the enrichment and preservation of oil and gas. The research results will provide important insights in the hydrocarbon exploration and exploitation in dolostone reservoirs of Tarim Basin.
•The conventional and image log response characteristics of different types of vertical karst zonation were summarized through cores, thin section and well log-calibration. The vertical karst zone of the Cambrian in the Tabei area of the Tarim Basin is clear, and can be divided into surfacekarst zone, vertical vadose zone, horizontal underflow zone and deep slow flow zone from up to bottom.•The conventional logs combined with image log method is used to clarify the typical logging response characteristics of the four karst zones to complete the rapid division of the exploration well karst zone in the study area.•By comparing the oil test and production data, it is clear that the vertical vadose zone is closely related to the oil and gas reservoir, which provides theoretical support for the exploration and development of oil and gas reservoirs.
Learning to Collaborate Feng, Jun; Li, Heng; Huang, Minlie ...
Proceedings of the 2018 World Wide Web Conference,
04/2018
Conference Proceeding
Open access
Ranking is a fundamental and widely studied problem in scenarios such as search, advertising, and recommendation. However, joint optimization for multi-scenario ranking, which aims to improve the ...overall performance of several ranking strategies in different scenarios, is rather untouched. Separately optimizing each individual strategy has two limitations. The first one is lack of collaboration between scenarios meaning that each strategy maximizes its own objective but ignores the goals of other strategies, leading to a sub-optimal overall performance. The second limitation is the inability of modeling the correlation between scenarios meaning that independent optimization in one scenario only uses its own user data but ignores the context in other scenarios. In this paper, we formulate multi-scenario ranking as a fully cooperative, partially observable, multi-agent sequential decision problem. We propose a novel model named Multi-Agent Recurrent Deterministic Policy Gradient (MA-RDPG) which has a communication component for passing messages, several private actors (agents) for making actions for ranking, and a centralized critic for evaluating the overall performance of the co-working actors. Each scenario is treated as an agent (actor). Agents collaborate with each other by sharing a global action-value function (the critic) and passing messages that encodes historical information across scenarios. The model is evaluated with online settings on a large E-commerce platform. Results show that the proposed model exhibits significant improvements against baselines in terms of the overall performance.
Recent advances have shown that symmetry, a structural prior that most objects exhibit, can support a variety of single-view 3D understanding tasks. However, detecting 3D symmetry from an image ...remains a challenging task. Previous works either assume that the symmetry is given or detect the symmetry with a heuristic-based method. In this paper, we present NeRD, a Neural 3D Reflection Symmetry Detector, which combines the strength of learning-based recognition and geometry-based reconstruction to accurately recover the normal direction of objects' mirror planes. Specifically, we first enumerate the symmetry planes with a coarse-to-fine strategy and then find the best ones by building 3D cost volumes to examine the intra-image pixel correspondence from the symmetry. Our experiments show that the symmetry planes detected with our method are significantly more accurate than the planes from direct CNN regression on both synthetic and real-world datasets. We also demonstrate that the detected symmetry can be used to improve the performance of downstream tasks such as pose estimation and depth map regression. The code of this paper has been made public at https://github.com/zhou13/nerd.
Cytoskeletal networks have a self-healing property where networks can repair defects to maintain structural integrity. However, both the mechanisms and dynamics of healing remain largely unknown. ...Here we report an unexplored healing mechanism in microtubule-motor networks by active crosslinking. We directly generate network cracks using a light-controlled microtubule-motor system, and observe that the cracks can self-heal. Combining theory and experiment, we find that the networks must overcome internal elastic resistance in order to heal cracks, giving rise to a bifurcation of dynamics dependent on the initial opening angle of the crack: the crack heals below a critical angle and opens up at larger angles. Simulation of a continuum model reproduces the bifurcation dynamics, revealing the importance of a boundary layer where free motors and microtubules can actively crosslink and thereby heal the crack. We also formulate a simple elastic-rod model that can qualitatively predict the critical angle, which is found to be tunable by two dimensionless geometric parameters, the ratio of the boundary layer and network width, and the aspect ratio of the network. Our results provide a new framework for understanding healing in cytoskeletal networks and designing self-healable biomaterials.
Synthesis of ergodic, stationary visual patterns is widely applicable in texturing, shape modeling, and digital content creation. The wide applicability of this technique thus requires the pattern ...synthesis approaches to be scalable, diverse, and authentic. In this paper, we propose an exemplar-based visual pattern synthesis framework that aims to model the inner statistics of visual patterns and generate new, versatile patterns that meet the aforementioned requirements. To this end, we propose an implicit network based on generative adversarial network (GAN) and periodic encoding, thus calling our network the Implicit Periodic Field Network (IPFN). The design of IPFN ensures scalability: the implicit formulation directly maps the input coordinates to features, which enables synthesis of arbitrary size and is computationally efficient for 3D shape synthesis. Learning with a periodic encoding scheme encourages diversity: the network is constrained to model the inner statistics of the exemplar based on spatial latent codes in a periodic field. Coupled with continuously designed GAN training procedures, IPFN is shown to synthesize tileable patterns with smooth transitions and local variations. Last but not least, thanks to both the adversarial training technique and the encoded Fourier features, IPFN learns high-frequency functions that produce authentic, high-quality results. To validate our approach, we present novel experimental results on various applications in 2D texture synthesis and 3D shape synthesis.
3D head avatars built with neural implicit volumetric representations have achieved unprecedented levels of photorealism. However, the computational cost of these methods remains a significant ...barrier to their widespread adoption, particularly in real-time applications such as virtual reality and teleconferencing. While attempts have been made to develop fast neural rendering approaches for static scenes, these methods cannot be simply employed to support realistic facial expressions, such as in the case of a dynamic facial performance. To address these challenges, we propose a novel fast 3D neural implicit head avatar model that achieves real-time rendering while maintaining fine-grained controllability and high rendering quality. Our key idea lies in the introduction of local hash table blendshapes, which are learned and attached to the vertices of an underlying face parametric model. These per-vertex hash-tables are linearly merged with weights predicted via a CNN, resulting in expression dependent embeddings. Our novel representation enables efficient density and color predictions using a lightweight MLP, which is further accelerated by a hierarchical nearest neighbor search method. Extensive experiments show that our approach runs in real-time while achieving comparable rendering quality to state-of-the-arts and decent results on challenging expressions.
3D reconstruction from a single RGB image is a challenging problem in computer vision. Previous methods are usually solely data-driven, which lead to inaccurate 3D shape recovery and limited ...generalization capability. In this work, we focus on object-level 3D reconstruction and present a geometry-based end-to-end deep learning framework that first detects the mirror plane of reflection symmetry that commonly exists in man-made objects and then predicts depth maps by finding the intra-image pixel-wise correspondence of the symmetry. Our method fully utilizes the geometric cues from symmetry during the test time by building plane-sweep cost volumes, a powerful tool that has been used in multi-view stereopsis. To our knowledge, this is the first work that uses the concept of cost volumes in the setting of single-image 3D reconstruction. We conduct extensive experiments on the ShapeNet dataset and find that our reconstruction method significantly outperforms the previous state-of-the-art single-view 3D reconstruction networks in term of the accuracy of camera poses and depth maps, without requiring objects being completely symmetric. Code is available at https://github.com/zhou13/symmetrynet.