arXiv subjects

Wei Wei

Publications and source records attributed to Wei Wei.

At least 469 records · Page 26Linked to original sources

arXiv2019

Reinforcement Learning Driven Heuristic Optimization

Heuristic algorithms such as simulated annealing, Concorde, and METIS are effective and widely used approaches to find solutions to combinatorial optimization problems. However, they are limited by the high sample complexity required to reach a reasonable solution from a cold-start. In this paper, we introduce a novel framework to generate better initial solutions for heuristic algorithms using reinforcement learning (RL), named RLHO. We augment the ability of heuristic algorithms to greedily improve upon an existing initial solution generated by RL, and demonstrate novel results where RL is able to leverage the performance of heuristics as a learning signal to generate better initialization. We apply this framework to Proximal Policy Optimization (PPO) and Simulated Annealing (SA). We conduct a series of experiments on the well-known NP-complete bin packing problem, and show that the RLHO method outperforms our baselines. We show that on the bin packing problem, RL can learn to help heuristics perform even better, allowing us to combine the best parts of both approaches.

cs.LG↗

arXiv2019

Distributed Optimal Load Frequency Control Considering Nonsmooth Cost Functions

This work addresses the distributed frequency control problem in power systems considering controllable load with a nonsmooth cost. The nonsmoothness exists widely in power systems, such as tiered price, greatly challenging the design of distributed optimal controllers. In this regard, we first formulate an optimization problem that minimizes the nonsmooth regulation cost, where both capacity limits of controllable load and tie-line flow are considered. Then, a distributed controller is derived using the Clark generalized gradient. We also prove the optimality of the equilibrium of the closed-loop system as well as its asymptotic stability. Simulations carried out on the IEEE 68-bus system verifies the effectiveness of the proposed method.

eess.SY↗

arXiv2019

Policy Certificates: Towards Accountable Reinforcement Learning

The performance of a reinforcement learning algorithm can vary drastically during learning because of exploration. Existing algorithms provide little information about the quality of their current policy before executing it, and thus have limited use in high-stakes applications like healthcare. We address this lack of accountability by proposing that algorithms output policy certificates. These certificates bound the sub-optimality and return of the policy in the next episode, allowing humans to intervene when the certified quality is not satisfactory. We further introduce two new algorithms with certificates and present a new framework for theoretical analysis that guarantees the quality of their policies and certificates. For tabular MDPs, we show that computing certificates can even improve the sample-efficiency of optimism-based exploration. As a result, one of our algorithms is the first to achieve minimax-optimal PAC bounds up to lower-order terms, and this algorithm also matches (and in some settings slightly improves upon) existing minimax regret bounds.

cs.LG↗

arXiv2019

Velocity Saturation effect on Low Frequency Noise in short channel Single Layer Graphene FETs

Graphene devices for analog and RF applications are prone to Low Frequency Noise (LFN) due to its upconversion to undesired phase noise at higher frequencies. Such applications demand the use of short channel graphene transistors that operate at high electric fields in order to ensure a high speed. Electric field is inversely proportional to device length and proportional to channel potential so it gets maximized as the drain voltage increases and the transistor length shrinks. Under these conditions though, short channel effects like Velocity Saturation (VS) should be taken into account. Carrier number and mobility fluctuations have been proved to be the main sources that generate LFN in graphene devices. While their contribution to the bias dependence of LFN in long channels has been thoroughly investigated, the way in which VS phenomenon affects LFN in short channel devices under high drain voltage conditions has not been well understood. At low electric field operation, VS effect is negligible since carriers velocity is far away from being saturated. Under these conditions, LFN can be precicely predicted by a recently established physics-based analytical model. The present paper goes a step furher and proposes a new model which deals with the contribution of VS effect on LFN under high electric field conditions. The implemented model is validated with novel experimental data, published for the first time, from CVD grown back-gated single layer graphene transistors operating at gigahertz frequencies. The model accurately captures the reduction of LFN especially near charge neutrality point because of the effect of VS mechanism. Moreover, an analytical expression for the effect of contact resistance on LFN is derived. This contact resistance contribution is experimentally shown to be dominant at higher gate voltages and is accurately described by the proposed model.

cond-mat.mes-hall↗

arXiv2019

InstaNAS: Instance-aware Neural Architecture Search

Conventional Neural Architecture Search (NAS) aims at finding a single architecture that achieves the best performance, which usually optimizes task related learning objectives such as accuracy. However, a single architecture may not be representative enough for the whole dataset with high diversity and variety. Intuitively, electing domain-expert architectures that are proficient in domain-specific features can further benefit architecture related objectives such as latency. In this paper, we propose InstaNAS---an instance-aware NAS framework---that employs a controller trained to search for a "distribution of architectures" instead of a single architecture; This allows the model to use sophisticated architectures for the difficult samples, which usually comes with large architecture related cost, and shallow architectures for those easy samples. During the inference phase, the controller assigns each of the unseen input samples with a domain expert architecture that can achieve high accuracy with customized inference costs. Experiments within a search space inspired by MobileNetV2 show InstaNAS can achieve up to 48.8% latency reduction without compromising accuracy on a series of datasets against MobileNetV2.

cs.LG↗

arXiv2019

Optoelectronic mixing with high frequency graphene transistors

Graphene is ideally suited for optoelectronic applications. It offers absorption at telecom wavelengths, high-frequency operation and CMOS-compatibility. We report optoelectronic mixing up to to 67GHz using a back-gated graphene field effect transistor (GFET). We also present a model to describe the resulting mixed current. These results pave the way for GETs optoelectronic mixers for mm-wave applications, such as telecommunications and RADAR/LIDAR systems.

physics.app-ph↗

arXiv2019

A Probabilistic Framework for Location Inference from Social Media

We study the extent to which we can infer users' geographical locations from social media. Location inference from social media can benefit many applications, such as disaster management, targeted advertising, and news content tailoring. The challenges, however, lie in the limited amount of labeled data and the large scale of social networks. In this paper, we formalize the problem of inferring location from social media into a semi-supervised factor graph model (SSFGM). The model provides a probabilistic framework in which various sources of information (e.g., content and social network) can be combined together. We design a two-layer neural network to learn feature representations, and incorporate the learned latent features into SSFGM. To deal with the large-scale problem, we propose a Two-Chain Sampling (TCS) algorithm to learn SSFGM. The algorithm achieves a good trade-off between accuracy and efficiency. Experiments on Twitter and Weibo show that the proposed TCS algorithm for SSFGM can substantially improve the inference accuracy over several state-of-the-art methods. More importantly, TCS achieves over 100x speedup comparing with traditional propagation-based methods (e.g., loopy belief propagation).

cs.AI↗

arXiv2019

$σ_2$ Yamabe problem on conic 4-spheres

We discuss the constant $σ_{2}$ problem for conic 4-spheres. Based on earlier works of Chang-Han-Yang and Han-Li-Teixeira, we are able to find a necessary condition for the existence problem. In particular, when the condition is sharp, we have the uniqueness result similar to that of Troyanov in dimension 2. It indicates that the boundary of the moduli of all conic 4-spheres with constant $σ_{2}$ metrics consists of conic spheres with 2 conic points and rotational symmetry.

math.DG↗

arXiv2019

Semi-supervised Transfer Learning for Image Rain Removal

Single image rain removal is a typical inverse problem in computer vision. The deep learning technique has been verified to be effective for this task and achieved state-of-the-art performance. However, previous deep learning methods need to pre-collect a large set of image pairs with/without synthesized rain for training, which tends to make the neural network be biased toward learning the specific patterns of the synthesized rain, while be less able to generalize to real test samples whose rain types differ from those in the training data. To this issue, this paper firstly proposes a semi-supervised learning paradigm toward this task. Different from traditional deep learning methods which only use supervised image pairs with/without synthesized rain, we further put real rainy images, without need of their clean ones, into the network training process. This is realized by elaborately formulating the residual between an input rainy image and its expected network output (clear image without rain) as a specific parametrized rain streaks distribution. The network is therefore trained to adapt real unsupervised diverse rain types through transferring from the supervised synthesized rain, and thus both the short-of-training-sample and bias-to-supervised-sample issues can be evidently alleviated. Experiments on synthetic and real data verify the superiority of our model compared to the state-of-the-arts.

cs.CV↗

arXiv2019

Vehicle Re-identification in Aerial Imagery: Dataset and Approach

In this work, we construct a large-scale dataset for vehicle re-identification (ReID), which contains 137k images of 13k vehicle instances captured by UAV-mounted cameras. To our knowledge, it is the largest UAV-based vehicle ReID dataset. To increase intra-class variation, each vehicle is captured by at least two UAVs at different locations, with diverse view-angles and flight-altitudes. We manually label a variety of vehicle attributes, including vehicle type, color, skylight, bumper, spare tire and luggage rack. Furthermore, for each vehicle image, the annotator is also required to mark the discriminative parts that helps them to distinguish this particular vehicle from others. Besides the dataset, we also design a specific vehicle ReID algorithm to make full use of the rich annotation information. It is capable of explicitly detecting discriminative parts for each specific vehicle and significantly outperforms the evaluated baselines and state-of-the-art vehicle ReID approaches.

cs.CV↗

arXiv2019

Pixel-aware Deep Function-mixture Network for Spectral Super-Resolution

Spectral super-resolution (SSR) aims at generating a hyperspectral image (HSI) from a given RGB image. Recently, a promising direction for SSR is to learn a complicated mapping function from the RGB image to the HSI counterpart using a deep convolutional neural network. This essentially involves mapping the RGB context within a size-specific receptive field centered at each pixel to its spectrum in the HSI. The focus thereon is to appropriately determine the receptive field size and establish the mapping function from RGB context to the corresponding spectrum. Due to their differences in category or spatial position, pixels in HSIs often require different-sized receptive fields and distinct mapping functions. However, few efforts have been invested to explicitly exploit this prior. To address this problem, we propose a pixel-aware deep function-mixture network for SSR, which is composed of a new class of modules, termed function-mixture (FM) blocks. Each FM block is equipped with some basis functions, i.e., parallel subnets of different-sized receptive fields. Besides, it incorporates an extra subnet as a mixing function to generate pixel-wise weights, and then linearly mixes the outputs of all basis functions with those generated weights. This enables us to pixel-wisely determine the receptive field size and the mapping function. Moreover, we stack several such FM blocks to further increase the flexibility of the network in learning the pixel-wise mapping. To encourage feature reuse, intermediate features generated by the FM blocks are fused in late stage, which proves to be effective for boosting the SSR performance. Experimental results on three benchmark HSI datasets demonstrate the superiority of the proposed method.

cs.CV↗

arXiv2019

On unconventional fall of rainbow spring

In this study we experimentally show that a stretched rainbow spring under gravity or extra weight may exhibit unconventional fall motion. Specially, when the rainbow spring is released from a high place, its lower end remains stationary until the spring wires stack together and then all the parts of the rainbow spring falls down together. We utilize a high-speed camera to record the fall process of one plastic and one metal rainbow spring under different loading conditions to systematically investigate this unconventional physical phenomenon. We use the time of the elastic wave propagating the length of the spring to predict the duration of the lower end of the rainbow spring remaining still. The findings from this study elucidate this physical phenomenon which has potential for the areas requiring temporary absolute space positioning.

physics.class-ph↗

arXiv2019

Complement Objective Training

Learning with a primary objective, such as softmax cross entropy for classification and sequence generation, has been the norm for training deep neural networks for years. Although being a widely-adopted approach, using cross entropy as the primary objective exploits mostly the information from the ground-truth class for maximizing data likelihood, and largely ignores information from the complement (incorrect) classes. We argue that, in addition to the primary objective, training also using a complement objective that leverages information from the complement classes can be effective in improving model performance. This motivates us to study a new training paradigm that maximizes the likelihood of the groundtruth class while neutralizing the probabilities of the complement classes. We conduct extensive experiments on multiple tasks ranging from computer vision to natural language understanding. The experimental results confirm that, compared to the conventional training with just one primary objective, training also with the complement objective further improves the performance of the state-of-the-art models across all tasks. In addition to the accuracy improvement, we also show that models trained with both primary and complement objectives are more robust to single-step adversarial attacks.

cs.LG↗

arXiv2019

Deep Learning for Multi-Messenger Astrophysics: A Gateway for Discovery in the Big Data Era

This report provides an overview of recent work that harnesses the Big Data Revolution and Large Scale Computing to address grand computational challenges in Multi-Messenger Astrophysics, with a particular emphasis on real-time discovery campaigns. Acknowledging the transdisciplinary nature of Multi-Messenger Astrophysics, this document has been prepared by members of the physics, astronomy, computer science, data science, software and cyberinfrastructure communities who attended the NSF-, DOE- and NVIDIA-funded "Deep Learning for Multi-Messenger Astrophysics: Real-time Discovery at Scale" workshop, hosted at the National Center for Supercomputing Applications, October 17-19, 2018. Highlights of this report include unanimous agreement that it is critical to accelerate the development and deployment of novel, signal-processing algorithms that use the synergy between artificial intelligence (AI) and high performance computing to maximize the potential for scientific discovery with Multi-Messenger Astrophysics. We discuss key aspects to realize this endeavor, namely (i) the design and exploitation of scalable and computationally efficient AI algorithms for Multi-Messenger Astrophysics; (ii) cyberinfrastructure requirements to numerically simulate astrophysical sources, and to process and interpret Multi-Messenger Astrophysics data; (iii) management of gravitational wave detections and triggers to enable electromagnetic and astro-particle follow-ups; (iv) a vision to harness future developments of machine and deep learning and cyberinfrastructure resources to cope with the scale of discovery in the Big Data Era; (v) and the need to build a community that brings domain experts together with data scientists on equal footing to maximize and accelerate discovery in the nascent field of Multi-Messenger Astrophysics.

astro-ph.IM↗

arXiv2018

Towards Effective Deep Embedding for Zero-Shot Learning

Zero-shot learning (ZSL) can be formulated as a cross-domain matching problem: after being projected into a joint embedding space, a visual sample will match against all candidate class-level semantic descriptions and be assigned to the nearest class. In this process, the embedding space underpins the success of such matching and is crucial for ZSL. In this paper, we conduct an in-depth study on the construction of embedding space for ZSL and posit that an ideal embedding space should satisfy two criteria: intra-class compactness and inter-class separability. While the former encourages the embeddings of visual samples of one class to distribute tightly close to the semantic description embedding of this class, the latter requires embeddings from different classes to be well separated from each other. Towards this goal, we present a simple but effective two-branch network to simultaneously map semantic descriptions and visual samples into a joint space, on which visual embeddings are forced to regress to their class-level semantic embeddings and the embeddings crossing classes are required to be distinguishable by a trainable classifier. Furthermore, we extend our method to a transductive setting to better handle the model bias problem in ZSL (i.e., samples from unseen classes tend to be categorized into seen classes) with minimal extra supervision. Specifically, we propose a pseudo labeling strategy to progressively incorporate the testing samples into the training process and thus balance the model between seen and unseen classes. Experimental results on five standard ZSL datasets show the superior performance of the proposed method and its transductive extension.

cs.CV↗

arXiv2018

MONAS: Multi-Objective Neural Architecture Search using Reinforcement Learning

Recent studies on neural architecture search have shown that automatically designed neural networks perform as good as expert-crafted architectures. While most existing works aim at finding architectures that optimize the prediction accuracy, these architectures may have complexity and is therefore not suitable being deployed on certain computing environment (e.g., with limited power budgets). We propose MONAS, a framework for Multi-Objective Neural Architectural Search that employs reward functions considering both prediction accuracy and other important objectives (e.g., power consumption) when searching for neural network architectures. Experimental results showed that, compared to the state-ofthe-arts, models found by MONAS achieve comparable or better classification accuracy on computer vision applications, while satisfying the additional objectives such as peak power.

cs.LG↗

arXiv2018

Miura-ori tube metamaterial with tunable dynamic property

Structures and/or materials with engineered functionality, capable of achieving targeted mechanical responses reacting to changes in external excitation, have various potential engineering applications, e.g. aerospace, oceanographic engineering, soft robot, and several others. Yet tunable mechanical performance is normally realized through carefully designing the architecture of structures, which is usually porous, leading to the complexity of the fabrication of the structures even using the recently emerged 3D printing technique. In this study we show that origami technique can provide an alternative solution to achieving the aim by carefully stacking the classical Miura sheets into the Miura-ori tube metamaterial and tuning the geometric parameters of the origami metamaterial. By combining numerical and experimental methods, we have demonstrated that an extremely broad range of natural frequency and dynamic response of the metamaterial can be achieved. The proposed structure can be easily fabricated from a single thin sheet made of one material and simultaneously owns better mechanical properties than the Miura sheet.

physics.app-ph↗

arXiv2018

Accurate Spectral Super-resolution from Single RGB Image Using Multi-scale CNN

Different from traditional hyperspectral super-resolution approaches that focus on improving the spatial resolution, spectral super-resolution aims at producing a high-resolution hyperspectral image from the RGB observation with super-resolution in spectral domain. However, it is challenging to accurately reconstruct a high-dimensional continuous spectrum from three discrete intensity values at each pixel, since too much information is lost during the procedure where the latent hyperspectral image is downsampled (e.g., with x10 scaling factor) in spectral domain to produce an RGB observation. To address this problem, we present a multi-scale deep convolutional neural network (CNN) to explicitly map the input RGB image into a hyperspectral image. Through symmetrically downsampling and upsampling the intermediate feature maps in a cascading paradigm, the local and non-local image information can be jointly encoded for spectral representation, ultimately improving the spectral reconstruction accuracy. Extensive experiments on a large hyperspectral dataset demonstrate the effectiveness of the proposed method.

cs.CV↗