arXiv subjects

Rui Yan

Publications and source records attributed to Rui Yan.

At least 253 records · Page 14Linked to original sources

arXiv2021

On the Detection of Markov Decision Processes

We study the detection problem for a finite set of Markov decision processes (MDPs) where the MDPs have the same state and action spaces but possibly different probabilistic transition functions. Any one of these MDPs could be the model for some underlying controlled stochastic process, but it is unknown a priori which MDP is the ground truth. We investigate whether it is possible to asymptotically detect the ground truth MDP model perfectly based on a single observed history (state-action sequence). Since the generation of histories depends on the policy adopted to control the MDPs, we discuss the existence and synthesis of policies that allow for perfect detection. We start with the case of two MDPs and establish a necessary and sufficient condition for the existence of policies that lead to perfect detection. Based on this condition, we then develop an algorithm that efficiently (in time polynomial in the size of the MDPs) determines the existence of policies and synthesizes one when they exist. We further extend the results to the more general case where there are more than two MDPs in the candidate set, and we develop a policy synthesis algorithm based on the breadth-first search and recursion. We demonstrate the effectiveness of our algorithms through numerical examples.

math.OC↗

arXiv2021

Video-Text Pre-training with Learned Regions

Video-Text pre-training aims at learning transferable representations from large-scale video-text pairs via aligning the semantics between visual and textual information. State-of-the-art approaches extract visual features from raw pixels in an end-to-end fashion. However, these methods operate at frame-level directly and thus overlook the spatio-temporal structure of objects in video, which yet has a strong synergy with nouns in textual descriptions. In this work, we propose a simple yet effective module for video-text representation learning, namely RegionLearner, which can take into account the structure of objects during pre-training on large-scale video-text pairs. Given a video, our module (1) first quantizes visual features into semantic clusters, then (2) generates learnable masks and uses them to aggregate the features belonging to the same semantic region, and finally (3) models the interactions between different aggregated regions. In contrast to using off-the-shelf object detectors, our proposed module does not require explicit supervision and is much more computationally efficient. We pre-train the proposed approach on the public WebVid2M and CC3M datasets. Extensive evaluations on four downstream video-text retrieval benchmarks clearly demonstrate the effectiveness of our RegionLearner. The code will be available at https://github.com/ruiyan1995/Region_Learner.

cs.CV↗

arXiv2021

TaskDrop: A Competitive Baseline for Continual Learning of Sentiment Classification

In this paper, we study the multi-task sentiment classification problem in the continual learning setting, i.e., a model is sequentially trained to classifier the sentiment of reviews of products in a particular category. The use of common sentiment words in reviews of different product categories leads to large cross-task similarity, which differentiates it from continual learning in other domains. This knowledge sharing nature renders forgetting reduction focused approaches less effective for the problem under consideration. Unlike existing approaches, where task-specific masks are learned with specifically presumed training objectives, we propose an approach called Task-aware Dropout (TaskDrop) to generate masks in a random way. While the standard dropout generates and applies random masks for each training instance per epoch for effective regularization, TaskDrop applies random masking for task-wise capacity allocation and reuse. We conducted experimental studies on three multi-task review datasets and made comparison to various baselines and state-of-the-art approaches. Our empirical results show that regardless of simplicity, TaskDrop overall achieved competitive performances for all the three datasets, especially after relative long term learning. This demonstrates that the proposed random capacity allocation mechanism works well for continual sentiment classification.

cs.CL↗

arXiv2021

Experimental Analysis of PandaX-4T Cryogenic Distillation System for Removing Krypton from Xenon

An efficient cryogenic distillation system was designed and constructed for PandaX-4T dark matter detector based on the McCabe-Thiele (M-T) method and the conservation of mass and energy. This distillation system is designed to reduce the concentration of krypton in commercial xenon from 5X$10^{-7}$ mol/mol to $10^{-14}$ mol/mol with 99% xenon collection efficiency at a maximum flow rate of 10 kg/h. The offline distillation operation has been completed and 5.75 tons of ultra-high purity xenon was produced, which is used as the detection medium in PandaX-4T detector. The krypton concentration of the product xenon is measured with an upper limit of 8.0 ppt. The stability and purification performance of the cryogenic distillation system are studied by analyzing the experimental data, which is important for theoretical research and distillation operation optimization.

physics.ins-det↗

arXiv2021

Research on Brick Schema Representation for Building Operation with Variable Refrigerant Flow Systems

Building metadata is regarded as the signpost in organizing massive building data. The application of building metadata simplifies the creation of digital representations and provides portable data analytics. Typical metadata standards such as Brick and Haystack are used to describe the data of the building system. Brick uses standard ontologies to create building metadata. However, neither Haystack nor Brick has provided definitions about the Variable Refrigerant Flow (VRF) system so far. For years, both Brick and Haystack working groups have been discussing how to describe VRF in their schema, mainly about the classification of VRF and the definitions of VRF units. There were no settled solutions for these problems. Meanwhile, the global VRF market is growing increasingly fast because of the energy efficiency and installation simplicity of the VRF system. It is needed to have the metadata to describe VRF units in buildings for data analysis and management. Addressing this challenge, this paper extended Brick Schema with the VRF module and verified the Brick VRF module. Then, the model and the service framework were developed and applied for a building in China. The framework can serve portable energy analysis for different areas. The VRF module of this paper provides a possible solution for the expression of the VRF system in the building semantic web. The works in this paper will support semantic web in automation strategies for building management and scalable building operation.

cs.OH↗

arXiv2021

Correlative image learning of chemo-mechanics in phase-transforming solids

Constitutive laws underlie most physical processes in nature. However, learning such equations in heterogeneous solids (e.g., due to phase separation) is challenging. One such relationship is between composition and eigenstrain, which governs the chemo-mechanical expansion in solids. In this work, we developed a generalizable, physically-constrained image-learning framework to algorithmically learn the chemo-mechanical constitutive law at the nanoscale from correlative four-dimensional scanning transmission electron microscopy and X-ray spectro-ptychography images. We demonstrated this approach on Li$_X$FePO$_4$, a technologically-relevant battery positive electrode material. We uncovered the functional form of composition-eigenstrain relation in this two-phase binary solid across the entire composition range (0 $\leq$ X $\leq$ 1), including inside the thermodynamically-unstable miscibility gap. The learned relation directly validates Vegard's law of linear response at the nanoscale. Our physics-constrained data-driven approach directly visualizes the residual strain field (by removing the compositional and coherency strain), which is otherwise impossible to quantify. Heterogeneities in the residual strain arise from misfit dislocations and were independently verified by X-ray diffraction line profile analysis. Our work provides the means to simultaneously quantify chemical expansion, coherency strain and dislocations in battery electrodes, which has implications on rate capabilities and lifetime. Broadly, this work also highlights the potential of integrating correlative microscopy and image learning for extracting material properties and physics.

cond-mat.mtrl-sci↗

arXiv2021

Multiplayer Homicidal Chauffeur Reach-Avoid Games via Guaranteed Winning Strategies

This paper studies a planar multiplayer Homicidal Chauffeur reach-avoid differential game, where each pursuer is a Dubins car and each evader has simple motion. The pursuers aim to protect a goal region cooperatively from the evaders. Due to the high-dimensional strategy space among pursuers, we decompose the whole game into multiple one-pursuer-one-evader subgames, each of which is solved in an analytical approach instead of solving Hamilton-Jacobi-Isaacs equations. For each subgame, an evasion region (ER) is introduced, based on which a pursuit strategy guaranteeing the winning of a simple-motion pursuer under specific conditions is proposed. Motivated by the simple-motion pursuer, a strategy for a Dubins-car pursuer is proposed when the pursuer-evader configuration satisfies separation condition (SC) and interception orientation (IO). The necessary and sufficient condition on capture radius, minimum turning radius and speed ratio to guarantee the pursuit winning is derived. When the IO is not satisfied (Non-IO), a heading adjustment pursuit strategy is proposed, and the condition to achieve IO within a finite time, is given. Then, a two-step pursuit strategy is proposed for the SC and Non-IO case. A non-convex optimization problem is introduced to give a condition guaranteeing the winning of the pursuer. A polynomial equation gives a lower bound of the non-convex problem, providing a sufficient and efficient pursuit winning condition. Finally, these pairwise outcomes are collected for the pursuer-evader matching. Simulations are provided to illustrate the theoretical results.

cs.GT↗

arXiv2021

Point Cloud Registration using Representative Overlapping Points

3D point cloud registration is a fundamental task in robotics and computer vision. Recently, many learning-based point cloud registration methods based on correspondences have emerged. However, these methods heavily rely on such correspondences and meet great challenges with partial overlap. In this paper, we propose ROPNet, a new deep learning model using Representative Overlapping Points with discriminative features for registration that transforms partial-to-partial registration into partial-to-complete registration. Specifically, we propose a context-guided module which uses an encoder to extract global features for predicting point overlap score. To better find representative overlapping points, we use the extracted global features for coarse alignment. Then, we introduce a Transformer to enrich point features and remove non-representative points based on point overlap score and feature matching. A similarity matrix is built in a partial-to-complete mode, and finally, weighted SVD is adopted to estimate a transformation matrix. Extensive experiments over ModelNet40 using noisy and partially overlapping point clouds show that the proposed method outperforms traditional and learning-based methods, achieving state-of-the-art performance. The code is available at https://github.com/zhulf0804/ROPNet.

cs.CV↗

arXiv2021

Privacy-Utility Trade-Offs Against Limited Adversaries

We study privacy-utility trade-offs where users share privacy-correlated useful information with a service provider to obtain some utility. The service provider is adversarial in the sense that it can infer the users' private information based on the shared useful information. To minimize the privacy leakage while maintaining a desired level of utility, the users carefully perturb the useful information via a probabilistic privacy mapping before sharing it. We focus on the setting in which the adversary attempting an inference attack on the users' privacy has potentially biased information about the statistical correlation between the private and useful variables. This information asymmetry between the users and the limited adversary leads to better privacy guarantees than the case of the omniscient adversary under the same utility requirement. We first identify assumptions on the adversary's information so that the inference costs are well-defined and finite. Then, we characterize the impact of the information asymmetry and show that it increases the inference costs for the adversary. We further formulate the design of the privacy mapping against a limited adversary using a difference of convex functions program and solve it via the concave-convex procedure. When the adversary's information is not precisely available, we adopt a Bayesian view and represent the adversary's information by a probability distribution. In this case, the expected cost for the adversary does not admit a closed-form expression, and we establish and maximize a lower bound of the expected cost. We provide a numerical example regarding a census data set to illustrate the theoretical results.

cs.IT↗

arXiv2021

Modeling Text-visual Mutual Dependency for Multi-modal Dialog Generation

Multi-modal dialog modeling is of growing interest. In this work, we propose frameworks to resolve a specific case of multi-modal dialog generation that better mimics multi-modal dialog generation in the real world, where each dialog turn is associated with the visual context in which it takes place. Specifically, we propose to model the mutual dependency between text-visual features, where the model not only needs to learn the probability of generating the next dialog utterance given preceding dialog utterances and visual contexts, but also the probability of predicting the visual features in which a dialog utterance takes place, leading the generated dialog utterance specific to the visual context. We observe significant performance boosts over vanilla models when the mutual dependency between text and visual features is modeled. Code is available at https://github.com/ShannonAI/OpenViDial.

cs.CL↗

arXiv2021

OpenViDial: A Large-Scale, Open-Domain Dialogue Dataset with Visual Contexts

When humans converse, what a speaker will say next significantly depends on what he sees. Unfortunately, existing dialogue models generate dialogue utterances only based on preceding textual contexts, and visual contexts are rarely considered. This is due to a lack of a large-scale multi-module dialogue dataset with utterances paired with visual contexts. In this paper, we release {\bf OpenViDial}, a large-scale multi-module dialogue dataset. The dialogue turns and visual contexts are extracted from movies and TV series, where each dialogue turn is paired with the corresponding visual context in which it takes place. OpenViDial contains a total number of 1.1 million dialogue turns, and thus 1.1 million visual contexts stored in images. Based on this dataset, we propose a family of encoder-decoder models leveraging both textual and visual contexts, from coarse-grained image features extracted from CNNs to fine-grained object features extracted from Faster R-CNNs. We observe that visual information significantly improves dialogue generation qualities, verifying the necessity of integrating multi-modal features for dialogue learning. Our work marks an important step towards large-scale multi-modal dialogue learning.

cs.CL↗

arXiv2021

Design and Commissioning of the PandaX-4T Cryogenic Distillation System for Krypton and Radon Removal

An online cryogenic distillation system for the removal of krypton and radon from xenon was designed and constructed for PandaX-4T, a highly sensitive dark matter detection experiment. The krypton content in a commercial xenon product is expected to be reduced by 7 orders of magnitude with 99% xenon collection efficiency at a flow rate of 10 kg/h by design. The same system can reduce radon content in xenon by reversed operation, with an expected radon reduction factor of about 1.8 in PandaX-4T under a flow rate of 56.5 kg/h. The commissioning of this system was completed, with krypton and radon operations tested under respective working conditions. The krypton concentration of the product xenon was measured with an upper limit of 8.0 ppt.

physics.ins-det↗

arXiv2021

Dialogue History Matters! Personalized Response Selectionin Multi-turn Retrieval-based Chatbots

Existing multi-turn context-response matching methods mainly concentrate on obtaining multi-level and multi-dimension representations and better interactions between context utterances and response. However, in real-place conversation scenarios, whether a response candidate is suitable not only counts on the given dialogue context but also other backgrounds, e.g., wording habits, user-specific dialogue history content. To fill the gap between these up-to-date methods and the real-world applications, we incorporate user-specific dialogue history into the response selection and propose a personalized hybrid matching network (PHMN). Our contributions are two-fold: 1) our model extracts personalized wording behaviors from user-specific dialogue history as extra matching information; 2) we perform hybrid representation learning on context-response utterances and explicitly incorporate a customized attention mechanism to extract vital information from context-response interactions so as to improve the accuracy of matching. We evaluate our model on two large datasets with user identification, i.e., personalized Ubuntu dialogue Corpus (P-Ubuntu) and personalized Weibo dataset (P-Weibo). Experimental results confirm that our method significantly outperforms several strong models by combining personalized attention, wording behaviors, and hybrid representation learning.

cs.CL↗

arXiv2021

How does Truth Evolve into Fake News? An Empirical Study of Fake News Evolution

Automatically identifying fake news from the Internet is a challenging problem in deception detection tasks. Online news is modified constantly during its propagation, e.g., malicious users distort the original truth and make up fake news. However, the continuous evolution process would generate unprecedented fake news and cheat the original model. We present the Fake News Evolution (FNE) dataset: a new dataset tracking the fake news evolution process. Our dataset is composed of 950 paired data, each of which consists of articles representing the three significant phases of the evolution process, which are the truth, the fake news, and the evolved fake news. We observe the features during the evolution and they are the disinformation techniques, text similarity, top 10 keywords, classification accuracy, parts of speech, and sentiment properties.

cs.CL↗

arXiv2021

Determination of responses of liquid xenon to low energy electron and nuclear recoils using the PandaX-II detector

We report a systematic determination of the responses of PandaX-II, a dual phase xenon time projection chamber detector, to low energy recoils. The electron recoil (ER) and nuclear recoil (NR) responses are calibrated, respectively, with injected tritiated methane or $^{220}$Rn source, and with $^{241}$Am-Be neutron source, within an energy range from $1-25$ keV (ER) and $4-80$ keV (NR), under the two drift fields of 400 and 317 V/cm. An empirical model is used to fit the light yield and charge yield for both types of recoils. The best fit models can well describe the calibration data. The systematic uncertainties of the fitted models are obtained via statistical comparison against the data.

physics.ins-det↗

arXiv2021

Neutronics Analysis for MSR Cell with Different Fuel Salt Channel Geometry

The neutronic properties of Molten Salt Reactor are different from that of traditional solid-fuel reactors due to its nuclear fuel particularity. Based upon MCNP code, the influence of the size and shape of fuel salt channel on neutron physics of MSR cell was studied systematically in this work. The results show that the infinite multiplication factors increases first and then decreases with the change of graphite cell size under the condition of given fuel volume fraction. In the case of the same FVF and average chord length, when the average chord length is relatively small, the k values with different fuel salt channel shapes are in good agreement; when the average chord length is relatively large, the k values with different fuel salt channel shapes are greatly different. In addition, some examples of practical application of this work are illustrated in the end, including cell selection for the core and thermal expansion displacement analysis of the cell.

physics.app-ph↗

arXiv2020

DSU-net: Dense SegU-net for automatic head-and-neck tumor segmentation in MR images

Precise and accurate segmentation of the most common head-and-neck tumor, nasopharyngeal carcinoma (NPC), in MRI sheds light on treatment and regulatory decisions making. However, the large variations in the lesion size and shape of NPC, boundary ambiguity, as well as the limited available annotated samples conspire NPC segmentation in MRI towards a challenging task. In this paper, we propose a Dense SegU-net (DSU-net) framework for automatic NPC segmentation in MRI. Our contribution is threefold. First, different from the traditional decoder in U-net using upconvolution for upsamling, we argue that the restoration from low resolution features to high resolution output should be capable of preserving information significant for precise boundary localization. Hence, we use unpooling to unsample and propose SegU-net. Second, to combat the potential vanishing-gradient problem, we introduce dense blocks which can facilitate feature propagation and reuse. Third, using only cross entropy (CE) as loss function may bring about troubles such as miss-prediction, therefore we propose to use a loss function comprised of both CE loss and Dice loss to train the network. Quantitative and qualitative comparisons are carried out extensively on in-house datasets, the experimental results show that our proposed architecture outperforms the existing state-of-the-art segmentation networks.

eess.IV↗

arXiv2020

Reasoning in Dialog: Improving Response Generation by Context Reading Comprehension

In multi-turn dialog, utterances do not always take the full form of sentences \cite{Carbonell1983DiscoursePA}, which naturally makes understanding the dialog context more difficult. However, it is essential to fully grasp the dialog context to generate a reasonable response. Hence, in this paper, we propose to improve the response generation performance by examining the model's ability to answer a reading comprehension question, where the question is focused on the omitted information in the dialog. Enlightened by the multi-task learning scheme, we propose a joint framework that unifies these two tasks, sharing the same encoder to extract the common and task-invariant features with different decoders to learn task-specific features. To better fusing information from the question and the dialog history in the encoding part, we propose to augment the Transformer architecture with a memory updater, which is designed to selectively store and update the history dialog information so as to support downstream tasks. For the experiment, we employ human annotators to write and examine a large-scale dialog reading comprehension dataset. Extensive experiments are conducted on this dataset, and the results show that the proposed model brings substantial improvements over several strong baselines on both tasks. In this way, we demonstrate that reasoning can indeed help better response generation and vice versa. We release our large-scale dataset for further research.

cs.CL↗