arXiv subjects

Yu Qian

Publications and source records attributed to Yu Qian.

28 records · Page 2Linked to original sources

arXiv2019

Can Sophisticated Dispatching Strategy Acquired by Reinforcement Learning? - A Case Study in Dynamic Courier Dispatching System

In this paper, we study a courier dispatching problem (CDP) raised from an online pickup-service platform of Alibaba. The CDP aims to assign a set of couriers to serve pickup requests with stochastic spatial and temporal arrival rate among urban regions. The objective is to maximize the revenue of served requests given a limited number of couriers over a period of time. Many online algorithms such as dynamic matching and vehicle routing strategy from existing literature could be applied to tackle this problem. However, these methods rely on appropriately predefined optimization objectives at each decision point, which is hard in dynamic situations. This paper formulates the CDP as a Markov decision process (MDP) and proposes a data-driven approach to derive the optimal dispatching rule-set under different scenarios. Our method stacks multi-layer images of the spatial-and-temporal map and apply multi-agent reinforcement learning (MARL) techniques to evolve dispatching models. This method solves the learning inefficiency caused by traditional centralized MDP modeling. Through comprehensive experiments on both artificial dataset and real-world dataset, we show: 1) By utilizing historical data and considering long-term revenue gains, MARL achieves better performance than myopic online algorithms; 2) MARL is able to construct the mapping between complex scenarios to sophisticated decisions such as the dispatching rule. 3) MARL has the scalability to adopt in large-scale real-world scenarios.

cs.AI↗

arXiv2019

A Multi-task Selected Learning Approach for Solving 3D Flexible Bin Packing Problem

A 3D flexible bin packing problem (3D-FBPP) arises from the process of warehouse packing in e-commerce. An online customer's order usually contains several items and needs to be packed as a whole before shipping. In particular, 5% of tens of millions of packages are using plastic wrapping as outer packaging every day, which brings pressure on the plastic surface minimization to save traditional logistics costs. Because of the huge practical significance, we focus on the issue of packing cuboid-shaped items orthogonally into a least-surface-area bin. The existing heuristic methods for classic 3D bin packing don't work well for this particular NP-hard problem and designing a good problem-specific heuristic is non-trivial. In this paper, rather than designing heuristics, we propose a novel multi-task framework based on Selected Learning to learn a heuristic-like policy that generates the sequence and orientations of items to be packed simultaneously. Through comprehensive experiments on a large scale real-world transaction order dataset and online AB tests, we show: 1) our selected learning method trades off the imbalance and correlation among the tasks and significantly outperforms the single task Pointer Network and the multi-task network without selected learning; 2) our method obtains an average 5.47% cost reduction than the well-designed greedy algorithm which is previously used in our online production system.

cs.LG↗

arXiv2018

Instance Retrieval at Fine-grained Level Using Multi-Attribute Recognition

In this paper, we present a method for instance ranking and retrieval at fine-grained level based on the global features extracted from a multi-attribute recognition model which is not dependent on landmarks information or part-based annotations. Further, we make this architecture suitable for mobile-device application by adopting the bilinear CNN to make the multi-attribute recognition model smaller (in terms of the number of parameters). The experiments run on the Dress category of DeepFashion In-Shop Clothes Retrieval and CUB200 datasets show that the results of instance retrieval at fine-grained level are promising for these datasets, specially in terms of texture and color.

cs.CV↗

arXiv2018

FineTag: Multi-attribute Classification at Fine-grained Level in Images

In this paper, we address the extraction of the fine-grained attributes of an instance as a `multi-attribute classification' problem. To this end, we propose an end-to-end architecture by adopting the bi-linear Convolutional Neural Network with the pairwise ranking loss. This is the first time such architecture is applied for the fine-grained attributes classification problem. We compared the proposed method with a competitive deep Convolutional Neural Network baseline. Extensive experiments show that the proposed method attains/outperforms the performance of compared baseline with significantly less number of parameters ($40\times$ less). We demonstrated our approach on CUB200 birds dataset whose annotations are adapted in this work for multi-attribute classification at fine-grained level.

cs.CV↗

arXiv2018

Improving the Annotation of DeepFashion Images for Fine-grained Attribute Recognition

DeepFashion is a widely used clothing dataset with 50 categories and more than overall 200k images where each image is annotated with fine-grained attributes. This dataset is often used for clothes recognition and although it provides comprehensive annotations, the attributes distribution is unbalanced and repetitive specially for training fine-grained attribute recognition models. In this work, we tailored DeepFashion for fine-grained attribute recognition task by focusing on each category separately. After selecting categories with sufficient number of images for training, we remove very scarce attributes and merge the duplicate ones in each category, then we clean the dataset based on the new list of attributes. We use a bilinear convolutional neural network with pairwise ranking loss function for multi-label fine-grained attribute recognition and show that the new annotations improve the results for such a task. The detailed annotations for each of the selected categories are provided for public use.

cs.CV↗

arXiv2017

Multi-kernel learning of deep convolutional features for action recognition

Image understanding using deep convolutional network has reached human-level performance, yet a closely related problem of video understanding especially, action recognition has not reached the requisite level of maturity. We combine multi-kernels based support-vector-machines (SVM) with a multi-stream deep convolutional neural network to achieve close to state-of-the-art performance on a 51-class activity recognition problem (HMDB-51 dataset); this specific dataset has proved to be particularly challenging for deep neural networks due to the heterogeneity in camera viewpoints, video quality, etc. The resulting architecture is named pillar networks as each (very) deep neural network acts as a pillar for the hierarchical classifiers. In addition, we illustrate that hand-crafted features such as improved dense trajectories (iDT) and Multi-skip Feature Stacking (MIFS), as additional pillars, can further supplement the performance.

cs.CV↗

arXiv2017

Pillar Networks++: Distributed non-parametric deep and wide networks

In recent work, it was shown that combining multi-kernel based support vector machines (SVMs) can lead to near state-of-the-art performance on an action recognition dataset (HMDB-51 dataset). This was 0.4\% lower than frameworks that used hand-crafted features in addition to the deep convolutional feature extractors. In the present work, we show that combining distributed Gaussian Processes with multi-stream deep convolutional neural networks (CNN) alleviate the need to augment a neural network with hand-crafted features. In contrast to prior work, we treat each deep neural convolutional network as an expert wherein the individual predictions (and their respective uncertainties) are combined into a Product of Experts (PoE) framework.

cs.CV↗

arXiv2015

A Muti-channel Distributed DAQ for n-TPC

A new fast neutron spectrometer named n-TPC has been designed by LPRI (Key Laboratory of Particle & Radiation Imaging, Ministry of Education) at Tsinghua University. The neutron energy spectrum can be calculated from the recoil angle and energy of the recoil proton detected by a 704-pad GEM-TPC. In beam tests at IHIP (Institute of Heavy Ion Physics, Peking University) in 2014, n-TPC performed better than 6 percents at 6MeV energy resolution and 0.5 percents detection efficiency. To find the best working parameters (the component and proportion of the gas, the high voltage between each GEM layer, etc.) of the n-TPC and support its application in various conditions, a multichannel distributed DAQ has been design to read out the signals from the 704 channels. With over 25 Ms/s sampling rate and 12 bit resolution for each channel, it can record the time and amplitude information as well as traditional DAQs in the TPC application domain. The main design objective of this distributed DAQ, however, is more flexible parameter modulation and operation. It can support the n-TPC without the limitation of the chassis and categorize signals arriving from the 704 channels at the same time by different events without event triggers.

physics.ins-det↗

arXiv2011

Universal driving structure of self-sustained oscillatory complex networks

Recently, self-sustained oscillations in complex networks consisting of nonoscillatory nodes (network oscillators) have attracted great interest in diverse natural and social fields. Due to complexity of network behaviors, little is known so far about the basic structures and fundamental rules underlying the oscillations, not to mention the principles of how to control it. In this article we propose a common design principle for oscillations; predict novel and universal Branched Circle (BC) structures of oscillatory networks based on this principle; and suggest an operable Complexity Reduction Method to reveal the BC structures. These ideas are applied to excitable cell networks (including neural cell networks), and genomic regulatory networks. Universal BC structures are identified clearly in these two considerably different systems. These BC structures reveal for the first time both oscillation sources and wave propagation pathways of complex networks, and guide us to control the oscillations with surprisingly high efficiency.

nlin.AO↗

arXiv2010

Structure and control of self-sustained target waves in excitable small-world networks

Small-world networks describe many important practical systems among which neural networks consisting of excitable nodes are the most typical ones. In this paper we study self-sustained oscillations of target waves in excitable small-world networks. A novel dominant phase-advanced driving (DPAD) method, which is generally applicable for analyzing all oscillatory complex networks consisting of nonoscillatory nodes, is proposed to reveal the self-organized structures supporting this type of oscillations. The DPAD method explicitly explores the oscillation sources and wave propagation paths of the systems, which are otherwise deeply hidden in the complicated patterns of randomly distributed target groups. Based on the understanding of the self-organized structure, the oscillatory patterns can be controlled with extremely high efficiency.

nlin.AO↗