arXiv subjects

Yong Yu

Publications and source records attributed to Yong Yu.

At least 271 records · Page 15Linked to original sources

arXiv2018

Time-resolved boson sampling with photons of different colors

Interference of multiple photons via a linear-optical network has profound applications for quantum foundation, quantum metrology and quantum computation. Particularly, a boson sampling experiment with a moderate number of photons becomes intractable even for the most powerful classical computers, and will lead to "quantum supremacy". Scaling up from small-scale experiments requires highly indistinguishable single photons, which may be prohibited for many physical systems. Here we experimentally demonstrate a time-resolved version of boson sampling by using photons not overlapping in their frequency spectra from three atomic-ensemble quantum memories. Time-resolved measurement enables us to observe nonclassical multiphoton correlation landscapes. An average fidelity over several interferometer configurations is measured to be 0.936(13), which is mainly limited by high-order events. Symmetries in the landscapes are identified to reflect symmetries of the optical network. Our work thus provides a route towards quantum supremacy with distinguishable photons.

quant-ph↗

arXiv2018

Bidding Machine: Learning to Bid for Directly Optimizing Profits in Display Advertising

Real-time bidding (RTB) based display advertising has become one of the key technological advances in computational advertising. RTB enables advertisers to buy individual ad impressions via an auction in real-time and facilitates the evaluation and the bidding of individual impressions across multiple advertisers. In RTB, the advertisers face three main challenges when optimizing their bidding strategies, namely (i) estimating the utility (e.g., conversions, clicks) of the ad impression, (ii) forecasting the market value (thus the cost) of the given ad impression, and (iii) deciding the optimal bid for the given auction based on the first two. Previous solutions assume the first two are solved before addressing the bid optimization problem. However, these challenges are strongly correlated and dealing with any individual problem independently may not be globally optimal. In this paper, we propose Bidding Machine, a comprehensive learning to bid framework, which consists of three optimizers dealing with each challenge above, and as a whole, jointly optimizes these three parts. We show that such a joint optimization would largely increase the campaign effectiveness and the profit. From the learning perspective, we show that the bidding machine can be updated smoothly with both offline periodical batch or online sequential training schemes. Our extensive offline empirical study and online A/B testing verify the high effectiveness of the proposed bidding machine.

cs.GT↗

arXiv2018

Wasserstein Distance Guided Representation Learning for Domain Adaptation

Domain adaptation aims at generalizing a high-performance learner on a target domain via utilizing the knowledge distilled from a source domain which has a different but related data distribution. One solution to domain adaptation is to learn domain invariant feature representations while the learned representations should also be discriminative in prediction. To learn such representations, domain adaptation frameworks usually include a domain invariant representation learning approach to measure and reduce the domain discrepancy, as well as a discriminator for classification. Inspired by Wasserstein GAN, in this paper we propose a novel approach to learn domain invariant feature representations, namely Wasserstein Distance Guided Representation Learning (WDGRL). WDGRL utilizes a neural network, denoted by the domain critic, to estimate empirical Wasserstein distance between the source and target samples and optimizes the feature extractor network to minimize the estimated Wasserstein distance in an adversarial manner. The theoretical advantages of Wasserstein distance for domain adaptation lie in its gradient property and promising generalization bound. Empirical studies on common sentiment and image classification adaptation datasets demonstrate that our proposed WDGRL outperforms the state-of-the-art domain invariant representation learning approaches.

stat.ML↗

arXiv2018

Collaborative Filtering with Graph-based Implicit Feedback

Introducing consumed items as users' implicit feedback in matrix factorization (MF) method, SVD++ is one of the most effective collaborative filtering methods for personalized recommender systems. Though powerful, SVD++ has two limitations: (i). only user-side implicit feedback is utilized, whereas item-side implicit feedback, which can also enrich item representations, is not leveraged;(ii). in SVD++, the interacted items are equally weighted when combining the implicit feedback, which can not reflect user's true preferences accurately. To tackle the above limitations, in this paper we propose Graph-based collaborative filtering (GCF) model, Weighted Graph-based collaborative filtering (W-GCF) model and Attentive Graph-based collaborative filtering (A-GCF) model, which (i). generalize the implicit feedback to item side based on the user-item bipartite graph; (ii). flexibly learn the weights of individuals in the implicit feedback hence improve the model's capacity. Comprehensive experiments show that our proposed models outperform state-of-the-art models.For sparse implicit feedback scenarios, additional improvement is further achieved by leveraging the step-two implicit feedback information.

cs.IR↗

arXiv2018

Unsupervised Deep Domain Adaptation for Pedestrian Detection

This paper addresses the problem of unsupervised domain adaptation on the task of pedestrian detection in crowded scenes. First, we utilize an iterative algorithm to iteratively select and auto-annotate positive pedestrian samples with high confidence as the training samples for the target domain. Meanwhile, we also reuse negative samples from the source domain to compensate for the imbalance between the amount of positive samples and negative samples. Second, based on the deep network we also design an unsupervised regularizer to mitigate influence from data noise. More specifically, we transform the last fully connected layer into two sub-layers - an element-wise multiply layer and a sum layer, and add the unsupervised regularizer to further improve the domain adaptation accuracy. In experiments for pedestrian detection, the proposed method boosts the recall value by nearly 30% while the precision stays almost the same. Furthermore, we perform our method on standard domain adaptation benchmarks on both supervised and unsupervised settings and also achieve state-of-the-art results.

cs.CV↗

arXiv2018

Texygen: A Benchmarking Platform for Text Generation Models

We introduce Texygen, a benchmarking platform to support research on open-domain text generation models. Texygen has not only implemented a majority of text generation models, but also covered a set of metrics that evaluate the diversity, the quality and the consistency of the generated texts. The Texygen platform could help standardize the research on text generation and facilitate the sharing of fine-tuned open-source implementations among researchers for their work. As a consequence, this would help in improving the reproductivity and reliability of future research work in text generation.

cs.CL↗

arXiv2017

Long Text Generation via Adversarial Training with Leaked Information

Automatically generating coherent and semantically meaningful text has many applications in machine translation, dialogue systems, image captioning, etc. Recently, by combining with policy gradient, Generative Adversarial Nets (GAN) that use a discriminative model to guide the training of the generative model as a reinforcement learning policy has shown promising results in text generation. However, the scalar guiding signal is only available after the entire text has been generated and lacks intermediate information about text structure during the generative process. As such, it limits its success when the length of the generated text samples is long (more than 20 words). In this paper, we propose a new framework, called LeakGAN, to address the problem for long text generation. We allow the discriminative net to leak its own high-level extracted features to the generative net to further help the guidance. The generator incorporates such informative signals into all generation steps through an additional Manager module, which takes the extracted features of current generated words and outputs a latent vector to guide the Worker module for next-word generation. Our extensive experiments on synthetic data and various real-world tasks with Turing test demonstrate that LeakGAN is highly effective in long text generation and also improves the performance in short text generation scenarios. More importantly, without any supervision, LeakGAN would be able to implicitly learn sentence structures only through the interaction between Manager and Worker.

cs.CL↗

arXiv2017

Supervised Hashing based on Energy Minimization

Recently, supervised hashing methods have attracted much attention since they can optimize retrieval speed and storage cost while preserving semantic information. Because hashing codes learning is NP-hard, many methods resort to some form of relaxation technique. But the performance of these methods can easily deteriorate due to the relaxation. Luckily, many supervised hashing formulations can be viewed as energy functions, hence solving hashing codes is equivalent to learning marginals in the corresponding conditional random field (CRF). By minimizing the KL divergence between a fully factorized distribution and the Gibbs distribution of this CRF, a set of consistency equations can be obtained, but updating them in parallel may not yield a local optimum since the variational lower bound is not guaranteed to increase. In this paper, we use a linear approximation of the sigmoid function to convert these consistency equations to linear systems, which have a closed-form solution. By applying this novel technique to two classical hashing formulations KSH and SPLH, we obtain two new methods called EM (energy minimizing based)-KSH and EM-SPLH. Experimental results on three datasets show the superiority of our methods.

cs.LG↗

arXiv2017

MAgent: A Many-Agent Reinforcement Learning Platform for Artificial Collective Intelligence

We introduce MAgent, a platform to support research and development of many-agent reinforcement learning. Unlike previous research platforms on single or multi-agent reinforcement learning, MAgent focuses on supporting the tasks and the applications that require hundreds to millions of agents. Within the interactions among a population of agents, it enables not only the study of learning algorithms for agents' optimal polices, but more importantly, the observation and understanding of individual agent's behaviors and social phenomena emerging from the AI society, including communication languages, leaderships, altruism. MAgent is highly scalable and can host up to one million agents on a single GPU server. MAgent also provides flexible configurations for AI researchers to design their customized environments and agents. In this demo, we present three environments designed on MAgent and show emerged collective intelligence by learning from scratch.

cs.LG↗

arXiv2017

Efficient Architecture Search by Network Transformation

Techniques for automatically designing deep neural network architectures such as reinforcement learning based approaches have recently shown promising results. However, their success is based on vast computational resources (e.g. hundreds of GPUs), making them difficult to be widely used. A noticeable limitation is that they still design and train each network from scratch during the exploration of the architecture space, which is highly inefficient. In this paper, we propose a new framework toward efficient architecture search by exploring the architecture space based on the current network and reusing its weights. We employ a reinforcement learning agent as the meta-controller, whose action is to grow the network depth or layer width with function-preserving transformations. As such, the previously validated networks can be reused for further exploration, thus saves a large amount of computational cost. We apply our method to explore the architecture space of the plain convolutional neural networks (no skip-connections, branching etc.) on image benchmark datasets (CIFAR-10, SVHN) with restricted computational resources (5 GPUs). Our method can design highly competitive networks that outperform existing networks using the same design scheme. On CIFAR-10, our model without skip-connections achieves 4.23\% test error rate, exceeding a vast majority of modern architectures and approaching DenseNet. Furthermore, by applying our method to explore the DenseNet architecture space, we are able to achieve more accurate networks with fewer parameters.

cs.LG↗

arXiv2017

Face Transfer with Generative Adversarial Network

Face transfer animates the facial performances of the character in the target video by a source actor. Traditional methods are typically based on face modeling. We propose an end-to-end face transfer method based on Generative Adversarial Network. Specifically, we leverage CycleGAN to generate the face image of the target character with the corresponding head pose and facial expression of the source. In order to improve the quality of generated videos, we adopt PatchGAN and explore the effect of different receptive field sizes on generated images.

cs.CV↗

arXiv2017

SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient

As a new way of training generative models, Generative Adversarial Nets (GAN) that uses a discriminative model to guide the training of the generative model has enjoyed considerable success in generating real-valued data. However, it has limitations when the goal is for generating sequences of discrete tokens. A major reason lies in that the discrete outputs from the generative model make it difficult to pass the gradient update from the discriminative model to the generative model. Also, the discriminative model can only assess a complete sequence, while for a partially generated sequence, it is non-trivial to balance its current score and the future one once the entire sequence has been generated. In this paper, we propose a sequence generation framework, called SeqGAN, to solve the problems. Modeling the data generator as a stochastic policy in reinforcement learning (RL), SeqGAN bypasses the generator differentiation problem by directly performing gradient policy update. The RL reward signal comes from the GAN discriminator judged on a complete sequence, and is passed back to the intermediate state-action steps using Monte Carlo search. Extensive experiments on synthetic data and real-world tasks demonstrate significant improvements over strong baselines.

cs.LG↗

arXiv2017

Unsupervised Diverse Colorization via Generative Adversarial Networks

Colorization of grayscale images has been a hot topic in computer vision. Previous research mainly focuses on producing a colored image to match the original one. However, since many colors share the same gray value, an input grayscale image could be diversely colored while maintaining its reality. In this paper, we design a novel solution for unsupervised diverse colorization. Specifically, we leverage conditional generative adversarial networks to model the distribution of real-world item colors, in which we develop a fully convolutional generator with multi-layer noise to enhance diversity, with multi-layer condition concatenation to maintain reality, and with stride 1 to keep spatial information. With such a novel network architecture, the model yields highly competitive performance on the open LSUN bedroom dataset. The Turing test of 80 humans further indicates our generated color schemes are highly convincible.

cs.CV↗

arXiv2017

Real-Time Bidding by Reinforcement Learning in Display Advertising

The majority of online display ads are served through real-time bidding (RTB) --- each ad display impression is auctioned off in real-time when it is just being generated from a user visit. To place an ad automatically and optimally, it is critical for advertisers to devise a learning algorithm to cleverly bid an ad impression in real-time. Most previous works consider the bid decision as a static optimization problem of either treating the value of each impression independently or setting a bid price to each segment of ad volume. However, the bidding for a given ad campaign would repeatedly happen during its life span before the budget runs out. As such, each bid is strategically correlated by the constrained budget and the overall effectiveness of the campaign (e.g., the rewards from generated clicks), which is only observed after the campaign has completed. Thus, it is of great interest to devise an optimal bidding strategy sequentially so that the campaign budget can be dynamically allocated across all the available impressions on the basis of both the immediate and future rewards. In this paper, we formulate the bid decision process as a reinforcement learning problem, where the state space is represented by the auction information and the campaign's real-time parameters, while an action is the bid price to set. By modeling the state transition via auction competition, we build a Markov Decision Process framework for learning the optimal bidding policy to optimize the advertising performance in the dynamic real-time bidding environment. Furthermore, the scalability problem from the large real-world auction volume and campaign budget is well handled by state value approximation using neural networks.

cs.LG↗

arXiv2016

Product-based Neural Networks for User Response Prediction

Predicting user responses, such as clicks and conversions, is of great importance and has found its usage in many Web applications including recommender systems, web search and online advertising. The data in those applications is mostly categorical and contains multiple fields; a typical representation is to transform it into a high-dimensional sparse binary feature representation via one-hot encoding. Facing with the extreme sparsity, traditional models may limit their capacity of mining shallow patterns from the data, i.e. low-order feature combinations. Deep models like deep neural networks, on the other hand, cannot be directly applied for the high-dimensional input because of the huge feature space. In this paper, we propose a Product-based Neural Networks (PNN) with an embedding layer to learn a distributed representation of the categorical data, a product layer to capture interactive patterns between inter-field categories, and further fully connected layers to explore high-order feature interactions. Our experimental results on two large-scale real-world ad click datasets demonstrate that PNNs consistently outperform the state-of-the-art models on various metrics.

cs.LG↗

arXiv2016

Twisted solutions to a simplified Ericksen-Leslie equation

In this article we construct global solutions to a simplified Ericksen-Leslie system on $\mathbb{R}^3$. The constructed solutions are twisted and periodic along the $x_3$-axis with period $d = 2π\big/ μ$. Here $μ> 0$ is the twist rate. $d$ is the distance between two planes which are parallel to the $x_1x_2$-plane. Liquid crystal material is placed in the region enclosed by these two planes. Given a well-prepared initial data, our solutions exist classically for all $t \in [0, \infty)$. However these solutions become singular at all points on the $x_3$-axis and escape into third dimension exponentially while $t \rightarrow \infty$. An optimal blow up rate is also obtained.

math.AP↗

arXiv2016

MeV Argon ion beam generation with narrow energy spread

Laser driven particle acceleration has shown remarkable progresses in generating multi-GeV electron bunches and 10s of MeV ion beams based on high-power laser facilities. Intense laser pulse offers the acceleration field of 1012 Volt per meter, several orders of magnitude larger than that in conventional accelerators, enabling compact devices. Here we report that a highly-collimated argon ion beam with narrow energy spread is produced by irradiating a 45-fs fully-relativistic laser pulse onto an argon cluster target. The highly-charged (Argon ion with charge state of 16+) heavy ion beam has a minimum absolute energy spread of 0.19 MeV per nucleon at the energy peak of 0.39 MeV per nucleon. we identify a novel scheme from particle-in-cell simulations that greatly reduces the beam energy spread. The laser-driven intense plasma wakefield has a strong modulation on the ion beam in a way that the low energy part is cut off. The pre-accelerated argon ion beam from Coulomb explosion thus becomes more mono-energetic and collimated.

physics.plasm-ph↗

arXiv2015

Differential measurement of atmospheric refraction with a telescope with double fields of view

For the sake of complete theoretical research of atmospheric refraction, the atmospheric refraction under the condition of lower angles of elevation is still worthy to be analyzed and explored. In some engineering applications, the objects with larger zenith distance must be observed sometimes. Carrying out observational research of the atmospheric refraction at lower angles of elevation has an important significance. It has been considered difficult to measure the atmospheric refraction at lower angles of elevation. A new idea for determining atmospheric refraction by utilizing differential measurement with double fields of view is proposed. Taking the observational principle of HIPPARCOS satellite as a reference, a schematic prototype with double fields of view was developed. In August of 2013, experimental observations were carried out and the atmospheric refractions at lower angles of elevation can be obtained by the schematic prototype. The measured value of the atmospheric refraction at the zenith distance of 78.8 degree is $240.23"\pm0.27"$, and the feasibility of differential measurement of atmospheric refraction with double fields of view was justified. The limitations of the schematic prototype such as inadequate ability of gathering light, lack of accurate meteorological data recording and lower automatic level of observation and data processing were also pointed out, which need to be improved in subsequent work.

astro-ph.IM↗