Paper Group ANR 185
Thermostatic control for demand response using distributed averaging and deep neural networks. Composable Core-sets for Determinant Maximization: A Simple Near-Optimal Algorithm. A Causal Perspective to Unbiased Conversion Rate Estimation on Data Missing Not at Random. Discriminative Clustering for Robust Unsupervised Domain Adaptation. Representat …
Thermostatic control for demand response using distributed averaging and deep neural networks
| Title | Thermostatic control for demand response using distributed averaging and deep neural networks |
| Authors | Kshitij Singh, Pratik K. Bajaria |
| Abstract | Smart buildings are the need of the day with increasing demand-supply ratios and deficiency to generate considerably. In any modern non-industrial infrastructure, these demands mainly comprise of thermostatically controlled loads (TCLs), which can be manoeuvred. TCL loads like air-conditioner, heater, refrigerator, are ubiquitous, and their operating times can be controlled to achieve desired aggregate power. This power aggregation, in turn, helps achieve load management targets and thereby serve as ancillary service (AS) to the power grid. In this work, a distributed averaging protocol is used to achieve the desired power aggregate set by the utility using steady-state desynchronization. The results are verified using a computer program for a homogeneous and heterogeneous population of TCLs. Further, load following scenario has been implemented using the utility as a reference. Apart from providing a significant AS to the power grid, the proposed idea also helps reduce the amplitude of power system oscillations imparted by the TCLs. Hardware-based results are obtained to verify its implementation feasibility in real-time. Additionally, we extend this idea to data-driven paradigm and provide comparisons therein. |
| Tasks | |
| Published | 2019-12-25 |
| URL | https://arxiv.org/abs/1912.11692v1 |
| https://arxiv.org/pdf/1912.11692v1.pdf | |
| PWC | https://paperswithcode.com/paper/thermostatic-control-for-demand-response |
| Repo | |
| Framework | |
Composable Core-sets for Determinant Maximization: A Simple Near-Optimal Algorithm
| Title | Composable Core-sets for Determinant Maximization: A Simple Near-Optimal Algorithm |
| Authors | Piotr Indyk, Sepideh Mahabadi, Shayan Oveis Gharan, Alireza Rezaei |
| Abstract | ``Composable core-sets’’ are an efficient framework for solving optimization problems in massive data models. In this work, we consider efficient construction of composable core-sets for the determinant maximization problem. This can also be cast as the MAP inference task for determinantal point processes, that have recently gained a lot of interest for modeling diversity and fairness. The problem was recently studied in [IMOR’18], where they designed composable core-sets with the optimal approximation bound of $\tilde O(k)^k$. On the other hand, the more practical Greedy algorithm has been previously used in similar contexts. In this work, first we provide a theoretical approximation guarantee of $O(C^{k^2})$ for the Greedy algorithm in the context of composable core-sets; Further, we propose to use a Local Search based algorithm that while being still practical, achieves a nearly optimal approximation bound of $O(k)^{2k}$; Finally, we implement all three algorithms and show the effectiveness of our proposed algorithm on standard data sets. | |
| Tasks | Point Processes |
| Published | 2019-07-06 |
| URL | https://arxiv.org/abs/1907.03197v1 |
| https://arxiv.org/pdf/1907.03197v1.pdf | |
| PWC | https://paperswithcode.com/paper/composable-core-sets-for-determinant-1 |
| Repo | |
| Framework | |
A Causal Perspective to Unbiased Conversion Rate Estimation on Data Missing Not at Random
| Title | A Causal Perspective to Unbiased Conversion Rate Estimation on Data Missing Not at Random |
| Authors | Wenhao Zhang, Wentian Bao, Xiao-Yang Liu, Keping Yang, Quan Lin, Hong Wen, Ramin Ramezani |
| Abstract | In modern e-commerce and advertising recommender systems, ongoing research works attempt to optimize conversion rate (CVR) estimation, and increase the gross merchandise volume. Even though the state-of-the-art CVR estimators adopt deep learning methods, their model performances are still subject to sample selection bias and data sparsity issues. Conversion labels of exposed items in training dataset are typically missing not at random due to selection bias. Empirically, data sparsity issue causes the performance degradation of model with large parameter space. In this paper, we proposed two causal estimators combined with multi-task learning, and aim to solve sample selection bias (SSB) and data sparsity (DS) issues in conversion rate estimation. The proposed estimators adjust for the MNAR mechanism as if they are trained on a “do dataset” where users are forced to click on all exposed items. We evaluate the causal estimators with billion data samples. Experiment results demonstrate that the proposed CVR estimators outperform other state-of-the-art CVR estimators. In addition, empirical study shows that our methods are cost-effective with large scale dataset. |
| Tasks | Multi-Task Learning, Recommendation Systems |
| Published | 2019-10-16 |
| URL | https://arxiv.org/abs/1910.09337v1 |
| https://arxiv.org/pdf/1910.09337v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-causal-perspective-to-unbiased-conversion |
| Repo | |
| Framework | |
Discriminative Clustering for Robust Unsupervised Domain Adaptation
| Title | Discriminative Clustering for Robust Unsupervised Domain Adaptation |
| Authors | Rui Wang, Guoyin Wang, Ricardo Henao |
| Abstract | Unsupervised domain adaptation seeks to learn an invariant and discriminative representation for an unlabeled target domain by leveraging the information of a labeled source dataset. We propose to improve the discriminative ability of the target domain representation by simultaneously learning tightly clustered target representations while encouraging that each cluster is assigned to a unique and different class from the source. This strategy alleviates the effects of negative transfer when combined with adversarial domain matching between source and target representations. Our approach is robust to differences in the source and target label distributions and thus applicable to both balanced and imbalanced domain adaptation tasks, and with a simple extension, it can also be used for partial domain adaptation. Experiments on several benchmark datasets for domain adaptation demonstrate that our approach can achieve state-of-the-art performance in all three scenarios, namely, balanced, imbalanced and partial domain adaptation. |
| Tasks | Domain Adaptation, Partial Domain Adaptation, Unsupervised Domain Adaptation |
| Published | 2019-05-30 |
| URL | https://arxiv.org/abs/1905.13331v1 |
| https://arxiv.org/pdf/1905.13331v1.pdf | |
| PWC | https://paperswithcode.com/paper/discriminative-clustering-for-robust |
| Repo | |
| Framework | |
Representation Degeneration Problem in Training Natural Language Generation Models
| Title | Representation Degeneration Problem in Training Natural Language Generation Models |
| Authors | Jun Gao, Di He, Xu Tan, Tao Qin, Liwei Wang, Tie-Yan Liu |
| Abstract | We study an interesting problem in training neural network-based models for natural language generation tasks, which we call the \emph{representation degeneration problem}. We observe that when training a model for natural language generation tasks through likelihood maximization with the weight tying trick, especially with big training datasets, most of the learnt word embeddings tend to degenerate and be distributed into a narrow cone, which largely limits the representation power of word embeddings. We analyze the conditions and causes of this problem and propose a novel regularization method to address it. Experiments on language modeling and machine translation show that our method can largely mitigate the representation degeneration problem and achieve better performance than baseline algorithms. |
| Tasks | Language Modelling, Machine Translation, Text Generation, Word Embeddings |
| Published | 2019-07-28 |
| URL | https://arxiv.org/abs/1907.12009v1 |
| https://arxiv.org/pdf/1907.12009v1.pdf | |
| PWC | https://paperswithcode.com/paper/representation-degeneration-problem-in-1 |
| Repo | |
| Framework | |
A Three-Player GAN: Generating Hard Samples To Improve Classification Networks
| Title | A Three-Player GAN: Generating Hard Samples To Improve Classification Networks |
| Authors | Simon Vandenhende, Bert De Brabandere, Davy Neven, Luc Van Gool |
| Abstract | We propose a Three-Player Generative Adversarial Network to improve classification networks. In addition to the game played between the discriminator and generator, a competition is introduced between the generator and the classifier. The generator’s objective is to synthesize samples that are both realistic and hard to label for the classifier. Even though we make no assumptions on the type of augmentations to learn, we find that the model is able to synthesize realistically looking examples that are hard for the classification model. Furthermore, the classifier becomes more robust when trained on these difficult samples. The method is evaluated on a public dataset for traffic sign recognition. |
| Tasks | Traffic Sign Recognition |
| Published | 2019-03-08 |
| URL | http://arxiv.org/abs/1903.03496v1 |
| http://arxiv.org/pdf/1903.03496v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-three-player-gan-generating-hard-samples-to |
| Repo | |
| Framework | |
CARL: Aggregated Search with Context-Aware Module Embedding Learning
| Title | CARL: Aggregated Search with Context-Aware Module Embedding Learning |
| Authors | Xinting Huang, Jianzhong Qi, Yu Sun, Rui Zhang, Hai-Tao Zheng |
| Abstract | Aggregated search aims to construct search result pages (SERPs) from blue-links and heterogeneous modules (such as news, images, and videos). Existing studies have largely ignored the correlations between blue-links and heterogeneous modules when selecting the heterogeneous modules to be presented. We observe that the top ranked blue-links, which we refer to as the \emph{context}, can provide important information about query intent and helps identify the relevant heterogeneous modules. For example, informative terms like “streamed” and “recorded” in the context imply that a video module may better satisfy the query. To model and utilize the context information for aggregated search, we propose a model with context attention and representation learning (CARL). Our model applies a recurrent neural network with an attention mechanism to encode the context, and incorporates the encoded context information into module embeddings. The context-aware module embeddings together with the ranking policy are jointly optimized under the Markov decision process (MDP) formulation. To achieve a more effective joint learning, we further propose an optimization function with self-supervision loss to provide auxiliary supervision signals. Experimental results based on two public datasets demonstrate the superiority of CARL over multiple baseline approaches, and confirm the effectiveness of the proposed optimization function in boosting the joint learning process. |
| Tasks | Representation Learning |
| Published | 2019-08-03 |
| URL | https://arxiv.org/abs/1908.03141v1 |
| https://arxiv.org/pdf/1908.03141v1.pdf | |
| PWC | https://paperswithcode.com/paper/carl-aggregated-search-with-context-aware |
| Repo | |
| Framework | |
Dual Illumination Estimation for Robust Exposure Correction
| Title | Dual Illumination Estimation for Robust Exposure Correction |
| Authors | Qing Zhang, Yongwei Nie, Wei-Shi Zheng |
| Abstract | Exposure correction is one of the fundamental tasks in image processing and computational photography. While various methods have been proposed, they either fail to produce visually pleasing results, or only work well for limited types of image (e.g., underexposed images). In this paper, we present a novel automatic exposure correction method, which is able to robustly produce high-quality results for images of various exposure conditions (e.g., underexposed, overexposed, and partially under- and over-exposed). At the core of our approach is the proposed dual illumination estimation, where we separately cast the under- and over-exposure correction as trivial illumination estimation of the input image and the inverted input image. By performing dual illumination estimation, we obtain two intermediate exposure correction results for the input image, with one fixes the underexposed regions and the other one restores the overexposed regions. A multi-exposure image fusion technique is then employed to adaptively blend the visually best exposed parts in the two intermediate exposure correction images and the input image into a globally well-exposed image. Experiments on a number of challenging images demonstrate the effectiveness of the proposed approach and its superiority over the state-of-the-art methods and popular automatic exposure correction tools. |
| Tasks | |
| Published | 2019-10-30 |
| URL | https://arxiv.org/abs/1910.13688v1 |
| https://arxiv.org/pdf/1910.13688v1.pdf | |
| PWC | https://paperswithcode.com/paper/dual-illumination-estimation-for-robust |
| Repo | |
| Framework | |
High Flux Passive Imaging with Single-Photon Sensors
| Title | High Flux Passive Imaging with Single-Photon Sensors |
| Authors | Atul Ingle, Andreas Velten, Mohit Gupta |
| Abstract | Single-photon avalanche diodes (SPADs) are an emerging technology with a unique capability of capturing individual photons with high timing precision. SPADs are being used in several active imaging systems (e.g., fluorescence lifetime microscopy and LiDAR), albeit mostly limited to low photon flux settings. We propose passive free-running SPAD (PF-SPAD) imaging, an imaging modality that uses SPADs for capturing 2D intensity images with unprecedented dynamic range under ambient lighting, without any active light source. Our key observation is that the precise inter-photon timing measured by a SPAD can be used for estimating scene brightness under ambient lighting conditions, even for very bright scenes. We develop a theoretical model for PF-SPAD imaging, and derive a scene brightness estimator based on the average time of darkness between successive photons detected by a PF-SPAD pixel. Our key insight is that due to the stochastic nature of photon arrivals, this estimator does not suffer from a hard saturation limit. Coupled with high sensitivity at low flux, this enables a PF-SPAD pixel to measure a wide range of scene brightness, from very low to very high, thereby achieving extreme dynamic range. We demonstrate an improvement of over 2 orders of magnitude over conventional sensors by imaging scenes spanning a dynamic range of 1,000,000:1. |
| Tasks | |
| Published | 2019-02-26 |
| URL | http://arxiv.org/abs/1902.10190v2 |
| http://arxiv.org/pdf/1902.10190v2.pdf | |
| PWC | https://paperswithcode.com/paper/high-flux-passive-imaging-with-single-photon |
| Repo | |
| Framework | |
Interpretable Apprenticeship Learning from Heterogeneous Decision-Making via Personalized Embeddings
| Title | Interpretable Apprenticeship Learning from Heterogeneous Decision-Making via Personalized Embeddings |
| Authors | Rohan Paleja, Andrew Silva, Letian Chen, Matthew Gombolay |
| Abstract | Advances in learning from demonstration (LfD) have enabled intelligent agents to learn decision-making strategies through observation. However, humans exhibit heterogeneity in their decision-making criteria, leading to demonstrations with significant variability. We propose a personalized apprenticeship learning framework that automatically infers an interpretable representation of all human task demonstrators by extracting latent, human-specific decision-making criteria specified by an inferred, personalized embedding. We achieve near-perfect LfD accuracy in synthetic domains and 89.02% accuracy on a real-world planning domain, significantly outperforming state-of-the-art benchmarks. Further, a user study conduced to assess the interpretability of different types of decision-making models finds evidence that our methodology produces both interpretable (p < 0.04) and highly usable models (p < 0.05). |
| Tasks | Decision Making |
| Published | 2019-06-14 |
| URL | https://arxiv.org/abs/1906.06397v2 |
| https://arxiv.org/pdf/1906.06397v2.pdf | |
| PWC | https://paperswithcode.com/paper/personalized-apprenticeship-learning-from |
| Repo | |
| Framework | |
Conditional Importance Sampling for Off-Policy Learning
| Title | Conditional Importance Sampling for Off-Policy Learning |
| Authors | Mark Rowland, Anna Harutyunyan, Hado van Hasselt, Diana Borsa, Tom Schaul, Rémi Munos, Will Dabney |
| Abstract | The principal contribution of this paper is a conceptual framework for off-policy reinforcement learning, based on conditional expectations of importance sampling ratios. This framework yields new perspectives and understanding of existing off-policy algorithms, and reveals a broad space of unexplored algorithms. We theoretically analyse this space, and concretely investigate several algorithms that arise from this framework. |
| Tasks | |
| Published | 2019-10-16 |
| URL | https://arxiv.org/abs/1910.07479v1 |
| https://arxiv.org/pdf/1910.07479v1.pdf | |
| PWC | https://paperswithcode.com/paper/conditional-importance-sampling-for-off |
| Repo | |
| Framework | |
Automated Real-time Anomaly Detection in Human Trajectories using Sequence to Sequence Networks
| Title | Automated Real-time Anomaly Detection in Human Trajectories using Sequence to Sequence Networks |
| Authors | Giorgos Bouritsas, Stelios Daveas, Antonios Danelakis, Constantinos Rizogiannis, Stelios C. A. Thomopoulos |
| Abstract | Detection of anomalous trajectories is an important problem with potential applications to various domains, such as video surveillance, risk assessment, vessel monitoring and high-energy physics. Modeling the distribution of trajectories with statistical approaches has been a challenging task due to the fact that such time series are usually non stationary and highly dimensional. However, modern machine learning techniques provide robust approaches for data-driven modeling and critical information extraction. In this paper, we propose a Sequence to Sequence architecture for real-time detection of anomalies in human trajectories, in the context of risk-based security. Our detection scheme is tested on a synthetic dataset of diverse and realistic trajectories generated by the ISL iCrowd simulator. The experimental results indicate that our scheme accurately detects motion patterns that deviate from normal behaviors and is promising for future real-world applications. |
| Tasks | Anomaly Detection, Time Series |
| Published | 2019-07-12 |
| URL | https://arxiv.org/abs/1907.05813v2 |
| https://arxiv.org/pdf/1907.05813v2.pdf | |
| PWC | https://paperswithcode.com/paper/automated-real-time-anomaly-detection-in |
| Repo | |
| Framework | |
Quantifying the Performance of Federated Transfer Learning
| Title | Quantifying the Performance of Federated Transfer Learning |
| Authors | Qinghe Jing, Weiyan Wang, Junxue Zhang, Han Tian, Kai Chen |
| Abstract | The scarcity of data and isolated data islands encourage different organizations to share data with each other to train machine learning models. However, there are increasing concerns on the problems of data privacy and security, which urges people to seek a solution like Federated Transfer Learning (FTL) to share training data without violating data privacy. FTL leverages transfer learning techniques to utilize data from different sources for training, while achieving data privacy protection without significant accuracy loss. However, the benefits come with a cost of extra computation and communication consumption, resulting in efficiency problems. In order to efficiently deploy and scale up FTL solutions in practice, we need a deep understanding on how the infrastructure affects the efficiency of FTL. Our paper tries to answer this question by quantitatively measuring a real-world FTL implementation FATE on Google Cloud. According to the results of carefully designed experiments, we verified that the following bottlenecks can be further optimized: 1) Inter-process communication is the major bottleneck; 2) Data encryption adds considerable computation overhead; 3) The Internet networking condition affects the performance a lot when the model is large. |
| Tasks | Transfer Learning |
| Published | 2019-12-30 |
| URL | https://arxiv.org/abs/1912.12795v1 |
| https://arxiv.org/pdf/1912.12795v1.pdf | |
| PWC | https://paperswithcode.com/paper/quantifying-the-performance-of-federated |
| Repo | |
| Framework | |
Relevance Feedback with Latent Variables in Riemann spaces
| Title | Relevance Feedback with Latent Variables in Riemann spaces |
| Authors | Simone Santini |
| Abstract | In this paper we develop and evaluate two methods for relevance feedback based on endowing a suitable “semantic query space” with a Riemann metric derived from the probability distribution of the positive samples of the feedback. The first method uses a Gaussian distribution to model the data, while the second uses a more complex Latent Semantic variable model. A mixed (discrete-continuous) version of the Expectation-Maximization algorithm is developed for this model. We motivate the need for the semantic query space by analyzing in some depth three well-known relevance feedback methods, and we develop a new experimental methodology to evaluate these methods and compare their performance in a neutral way, that is, without making assumptions on the system in which they will be embedded. |
| Tasks | |
| Published | 2019-06-15 |
| URL | https://arxiv.org/abs/1906.06526v1 |
| https://arxiv.org/pdf/1906.06526v1.pdf | |
| PWC | https://paperswithcode.com/paper/relevance-feedback-with-latent-variables-in |
| Repo | |
| Framework | |
Context-Based Dynamic Pricing with Online Clustering
| Title | Context-Based Dynamic Pricing with Online Clustering |
| Authors | Sentao Miao, Xi Chen, Xiuli Chao, Jiaxi Liu, Yidong Zhang |
| Abstract | We consider a context-based dynamic pricing problem of online products which have low sales. Sales data from Alibaba, a major global online retailer, illustrate the prevalence of low-sale products. For these products, existing single-product dynamic pricing algorithms do not work well due to insufficient data samples. To address this challenge, we propose pricing policies that concurrently perform clustering over products and set individual pricing decisions on the fly. By clustering data and identifying products that have similar demand patterns, we utilize sales data from products within the same cluster to improve demand estimation and allow for better pricing decisions. We evaluate the algorithms using the regret, and the result shows that when product demand functions come from multiple clusters, our algorithms significantly outperform traditional single-product pricing policies. Numerical experiments using a real dataset from Alibaba demonstrate that the proposed policies, compared with several benchmark policies, increase the revenue. The results show that online clustering is an effective approach to tackling dynamic pricing problems associated with low-sale products. Our algorithms were further implemented in a field study at Alibaba with 40 products for 30 consecutive days, and compared to the products which use business-as-usual pricing policy of Alibaba. The results from the field experiment show that the overall revenue increased by 10.14%. |
| Tasks | |
| Published | 2019-02-17 |
| URL | https://arxiv.org/abs/1902.06199v2 |
| https://arxiv.org/pdf/1902.06199v2.pdf | |
| PWC | https://paperswithcode.com/paper/context-based-dynamic-pricing-with-online |
| Repo | |
| Framework | |
