Guan Wang

Guan Wang, Hamid Laga, Anuj Srivastava

How can one analyze detailed 3D biological objects, such as neurons and botanical trees, that exhibit complex geometrical and topological variation? In this paper, we develop a novel mathematical framework for representing, comparing, and computing geodesic deformations between the shapes of such tree-like 3D objects. A hierarchical organization of subtrees characterizes these objects -- each subtree has the main branch with some side branches attached -- and one needs to match these structures across objects for meaningful comparisons. We propose a novel representation that extends the Square-Root Velocity Function (SRVF), initially developed for Euclidean curves, to tree-shaped 3D objects. We then define a new metric that quantifies the bending, stretching, and branch sliding needed to deform one tree-shaped object into the other. Compared to the current metrics, such as the Quotient Euclidean Distance (QED) and the Tree Edit Distance (TED), the proposed representation and metric capture the full elasticity of the branches (i.e., bending and stretching) as well as the topological variations (i.e., branch death/birth and sliding). It completely avoids the shrinkage that results from the edge collapse and node split operations of the QED and TED metrics. We demonstrate the utility of this framework in comparing, matching, and computing geodesics between biological objects such as neurons and botanical trees. The framework is also applied to various shape analysis tasks: (i) symmetry analysis and symmetrization of tree-shaped 3D objects, (ii) computing summary statistics (means and modes of variations) of populations of tree-shaped 3D objects, (iii) fitting parametric probability distributions to such populations, and (iv) finally synthesizing novel tree-shaped 3D objects through random sampling from estimated probability distributions.

Yuyang Zhang, Dik Hin Leung, Min Guo et al.

Matrix multiplication is the bedrock in Deep Learning inference application. When it comes to hardware acceleration on edge computing devices, matrix multiplication often takes up a great majority of the time. To achieve better performance in edge computing, we introduce a low-power Multi-layer Perceptron (MLP) accelerator based on a pipelined matrix multiplication scheme and a nonuniform quantization methodology. The implementation is running on Field-programmable Gate Array (FPGA) devices and tested its performance on handwritten digit classification and Q-learning tasks. Results show that our method can achieve better performance with fewer power consumption.

Biye Jiang, Pengye Zhang, Rihan Chen et al.

Modern large-scale systems such as recommender system and online advertising system are built upon computation-intensive infrastructure. The typical objective in these applications is to maximize the total revenue, e.g. GMV~(Gross Merchandise Volume), under a limited computation resource. Usually, the online serving system follows a multi-stage cascade architecture, which consists of several stages including retrieval, pre-ranking, ranking, etc. These stages usually allocate resource manually with specific computing power budgets, which requires the serving configuration to adapt accordingly. As a result, the existing system easily falls into suboptimal solutions with respect to maximizing the total revenue. The limitation is due to the face that, although the value of traffic requests vary greatly, online serving system still spends equal computing power among them. In this paper, we introduce a novel idea that online serving system could treat each traffic request differently and allocate "personalized" computation resource based on its value. We formulate this resource allocation problem as a knapsack problem and propose a Dynamic Computation Allocation Framework~(DCAF). Under some general assumptions, DCAF can theoretically guarantee that the system can maximize the total revenue within given computation budget. DCAF brings significant improvement and has been deployed in the display advertising system of Taobao for serving the main traffic. With DCAF, we are able to maintain the same business performance with 20\% computation resource reduction.

James MacGlashan, Mark K Ho, Robert Loftin et al.

This paper investigates the problem of interactively learning behaviors communicated by a human teacher using positive and negative feedback. Much previous work on this problem has made the assumption that people provide feedback for decisions that is dependent on the behavior they are teaching and is independent from the learner's current policy. We present empirical results that show this assumption to be false -- whether human trainers give a positive or negative feedback for a decision is influenced by the learner's current policy. Based on this insight, we introduce {\em Convergent Actor-Critic by Humans} (COACH), an algorithm for learning from policy-dependent feedback that converges to a local optimum. Finally, we demonstrate that COACH can successfully learn multiple behaviors on a physical robot.

Qingbo Hu, Guan Wang, Philip S. Yu

Online video websites receive huge amount of videos daily from users all around the world. How to provide valuable recommendations to viewers is an important task for both video websites and related third parties, such as search engines. Previous work conducted numerous analysis on the view counts of videos, which measure a video's value in terms of popularity. However, the long-lasting value of an online video, namely longevity, is hidden behind the history that a video accumulates its "popularity" through time. Generally speaking, a longevous video tends to constantly draw society's attention. With focus on one of the leading video websites, Youtube, this paper proposes a scoring mechanism quantifying a video's longevity. Evaluating a video's longevity can not only improve a video recommender system, but also help us to discover videos having greater advertising value, as well as adjust a video website's strategy of storing videos to shorten its responding time. In order to accurately quantify longevity, we introduce the concept of latent social impulses and how to use them measure a video's longevity. In order to derive latent social impulses, we view the video website as a digital signal filter and formulate the task as a convex minimization problem. The proposed longevity computation is based on the derived social impulses. Unfortunately, the required information to derive social impulses are not always public, which makes a third party unable to directly evaluate every video's longevity. To solve this problem, we formulate a semi-supervised learning task by using part of videos having known longevity scores to predict the unknown longevity scores. We propose a Gaussian Random Markov model with Loopy Belief Propagation to solve this problem. The conducted experiments on Youtube demonstrate that the proposed method significantly improves the prediction results comparing to baselines.