arXiv subjects

Peng Xu

Publications and source records attributed to Peng Xu.

At least 271 records · Page 15Linked to original sources

arXiv2022

Charactering instrumental noises and stochastic gravitational wave signals from combined time-delay interferometry

LISA will detect gravitational waves (GWs) in the milli-Hz frequency band in space. Time-delay interferometry (TDI) is developed to suppress laser frequency noise beneath the acceleration noise and optical metrology noise. To identify stochastic GW signals, it would be required to characterize these noise components entangled in TDI data streams. In this work, we investigate noises characterization by combining the first-generation TDI channels from Michelson and Relay configurations. The Michelson channels are helpful to characterize acceleration noises in the lower frequency band, and the Relay configuration could effectively resolve optical path noises in the higher frequencies. Synergy could be achieved from their combination to determine these instrumental noises. Based on the characterized noises, we further reconstruct the power spectrum of noise in the selected TDI channel. Two cases are performed to characterize the spectrum shape of a stochastic GW signal. For a modeled signal, its parameter(s) could be directly estimated from the TDI data, and its spectrum could be recovered from the inferred values. And for an unexpected signal, its spectrum may be recognized and retrieved from noise-subtracted residual in which its power spectral density surpasses the noise level.

gr-qc↗

arXiv2022

Global Gravity Field Model from Taiji-1 Observations

Taiji-1 is the first technology demonstration satellite of the Taiji program of China's space-borne gravitational wave antenna. After the demonstration of the key individual technologies, Taiji-1 continues collecting the data of the precision orbit determinations, satellite attitudes, and non-conservative forces exerted on the S/C. Therefore, during its free-fall, Taiji-1 can be viewed as operating in the high-low satellite-to-satellite tracking mode of a gravity recovery mission. In this work, we have selected and analyzed the one month data from Taiji-1's observations, and developed the techniques to resolve the long term interruptions and disturbances in the data due to the scheduled technology demonstration experiments. The first global gravity model \texttt{TJGM-r1911}, that independently derived from China's own satellite mission, is successfully built from Taiji-1's observations. Compared with gravity models from CHAMP and other satellite gravity missions, the accuracy discrepancies exist, which is mainly caused by the data discontinuity problem. As the extended free-falling phase been approved, Taiji-1 could serve as a gravity recovery mission for China since 2022 and it will provide us the independent measurement of both the static and the monthly time-variable global gravity field.

gr-qc↗

arXiv2022

Do As I Can, Not As I Say: Grounding Language in Robotic Affordances

Large language models can encode a wealth of semantic knowledge about the world. Such knowledge could be extremely useful to robots aiming to act upon high-level, temporally extended instructions expressed in natural language. However, a significant weakness of language models is that they lack real-world experience, which makes it difficult to leverage them for decision making within a given embodiment. For example, asking a language model to describe how to clean a spill might result in a reasonable narrative, but it may not be applicable to a particular agent, such as a robot, that needs to perform this task in a particular environment. We propose to provide real-world grounding by means of pretrained skills, which are used to constrain the model to propose natural language actions that are both feasible and contextually appropriate. The robot can act as the language model's "hands and eyes," while the language model supplies high-level semantic knowledge about the task. We show how low-level skills can be combined with large language models so that the language model provides high-level knowledge about the procedures for performing complex and temporally-extended instructions, while value functions associated with these skills provide the grounding necessary to connect this knowledge to a particular physical environment. We evaluate our method on a number of real-world robotic tasks, where we show the need for real-world grounding and that this approach is capable of completing long-horizon, abstract, natural language instructions on a mobile manipulator. The project's website and the video can be found at https://say-can.github.io/.

cs.RO↗

arXiv2022

Entailment Tree Explanations via Iterative Retrieval-Generation Reasoner

Large language models have achieved high performance on various question answering (QA) benchmarks, but the explainability of their output remains elusive. Structured explanations, called entailment trees, were recently suggested as a way to explain and inspect a QA system's answer. In order to better generate such entailment trees, we propose an architecture called Iterative Retrieval-Generation Reasoner (IRGR). Our model is able to explain a given hypothesis by systematically generating a step-by-step explanation from textual premises. The IRGR model iteratively searches for suitable premises, constructing a single entailment step at a time. Contrary to previous approaches, our method combines generation steps and retrieval of premises, allowing the model to leverage intermediate conclusions, and mitigating the input size limit of baseline encoder-decoder models. We conduct experiments using the EntailmentBank dataset, where we outperform existing benchmarks on premise retrieval and entailment tree generation, with around 300% gain in overall correctness.

cs.CL↗

arXiv2022

Density functional theory plus dynamical mean field theory within the framework of linear combination of numerical atomic orbitals: Formulation and benchmarks

The combination of density functional theory with dynamical mean-field theory (DFT+DMFT) has become a powerful first-principles approach to tackle strongly correlated materials in condensed matter physics. The wide use of this approach relies on robust and easy-to-use implementations, and its implementation in various numerical frameworks will increase its applicability on the one hand and help crosscheck the validity of the obtained results on the other. In the work, we develop a formalism within the linear combination of numerical atomic orbital (NAO) basis set framework, which allows for merging NAO-based DFT codes with DMFT quantum impurity solvers. The formalism is implemented by interfacing two NAO-based DFT codes with three DMFT impurity solvers, and its validity is testified by benchmark calculations for a wide range of strongly correlated materials, including 3\textit{d} transition metal compounds, lanthanides, and actinides. Our work not only enables DFT+DMFT calculations using popular and rapidly developing NAO-based DFT code packages, but also facilitates the combination of more advanced beyond-DFT methodologies available in this codes with the DMFT machinery.

cond-mat.str-el↗

arXiv2022

Newton-MR: Inexact Newton Method With Minimum Residual Sub-problem Solver

We consider a variant of inexact Newton Method, called Newton-MR, in which the least-squares sub-problems are solved approximately using Minimum Residual method. By construction, Newton-MR can be readily applied for unconstrained optimization of a class of non-convex problems known as invex, which subsumes convexity as a sub-class. For invex optimization, instead of the classical Lipschitz continuity assumptions on gradient and Hessian, Newton-MR's global convergence can be guaranteed under a weaker notion of joint regularity of Hessian and gradient. We also obtain Newton-MR's problem-independent local convergence to the set of minima. We show that fast local/global convergence can be guaranteed under a novel inexactness condition, which, to our knowledge, is much weaker than the prior related works. Numerical results demonstrate the performance of Newton-MR as compared with several other Newton-type alternatives on a few machine learning problems.

math.OC↗

arXiv2022

ASCEND: A Spontaneous Chinese-English Dataset for Code-switching in Multi-turn Conversation

Code-switching is a speech phenomenon occurring when a speaker switches language during a conversation. Despite the spontaneous nature of code-switching in conversational spoken language, most existing works collect code-switching data from read speech instead of spontaneous speech. ASCEND (A Spontaneous Chinese-English Dataset) is a high-quality Mandarin Chinese-English code-switching corpus built on spontaneous multi-turn conversational dialogue sources collected in Hong Kong. We report ASCEND's design and procedure for collecting the speech data, including annotations. ASCEND consists of 10.62 hours of clean speech, collected from 23 bilingual speakers of Chinese and English. Furthermore, we conduct baseline experiments using pre-trained wav2vec 2.0 models, achieving a best performance of 22.69\% character error rate and 27.05% mixed error rate.

cs.CL↗

arXiv2022

Tracing Astrophysical Black Hole Seeds and Primordial Black Holes with LISA-Taiji Network

In this work, we discuss the improvement that the joint network of LISA and Taiji could provide on exploring two kinds of black hole formation mechanisms. For astrophysical origin, we consider light seed and heavy seed scenarios, and generate populations accordingly. We find that the joint network has the potential to observe growing light seeds in the range 15 < z < 20, while single detector can hardly see, which would shed light on the light seeding mechanism. For the heavy seeds, the joint network only improves the signal-to-noise ratio. For primordial origin, we calculate the event rate at z > 20 and detection rates of LISA and the joint network. The joint network expands LISA's horizon towards lower mass end, where the event rate is high, so we have better chance observing primordial black holes with the joint network. We also estimate the parameters using Fisher matrices of LISA and the joint network, and find that the joint network significantly improves the estimation.

gr-qc↗

arXiv2022

TOD-CNN: An Effective Convolutional Neural Network for Tiny Object Detection in Sperm Videos

The detection of tiny objects in microscopic videos is a problematic point, especially in large-scale experiments. For tiny objects (such as sperms) in microscopic videos, current detection methods face challenges in fuzzy, irregular, and precise positioning of objects. In contrast, we present a convolutional neural network for tiny object detection (TOD-CNN) with an underlying data set of high-quality sperm microscopic videos (111 videos, $>$ 278,000 annotated objects), and a graphical user interface (GUI) is designed to employ and test the proposed model effectively. TOD-CNN is highly accurate, achieving $85.60\%$ AP$_{50}$ in the task of real-time sperm detection in microscopic videos. To demonstrate the importance of sperm detection technology in sperm quality analysis, we carry out relevant sperm quality evaluation metrics and compare them with the diagnosis results from medical doctors.

cs.CV↗

arXiv2022

Inexact Newton-CG Algorithms With Complexity Guarantees

We consider variants of a recently-developed Newton-CG algorithm for nonconvex problems \citep{royer2018newton} in which inexact estimates of the gradient and the Hessian information are used for various steps. Under certain conditions on the inexactness measures, we derive iteration complexity bounds for achieving $ε$-approximate second-order optimality that match best-known lower bounds. Our inexactness condition on the gradient is adaptive, allowing for crude accuracy in regions with large gradients. We describe two variants of our approach, one in which the step-size along the computed search direction is chosen adaptively and another in which the step-size is pre-defined. To obtain second-order optimality, our algorithms will make use of a negative curvature direction on some steps. These directions can be obtained, with high-probability, using a certain randomized algorithm. In this sense, all of our results hold with high-probability over the run of the algorithm. We evaluate the performance of our proposed algorithms empirically on several machine learning models.

math.OC↗

arXiv2022

$δ$-$r$-Hyperideals and $ϕ$-$δ$-$r$-Hyperideals of Commutative Krasner Hyperrings

In this paper, our purpose is to define the expansion of $r$-hyperideals and extend this concept to $ϕ$-$δ$-$r$-hyperideal. Let $\Re$ be a commutative Krasner hyperring with nonzero identity. Given an expansion $δ$ of hyperideals, a proper hyperideal $N$ of $\Re$ is called $δ$-$r$-hyperideal if $a\cdot b\in N$ with $ann(a)=0$ implies that $b\in δ(N)$, for all $a,b\in\Re$. Therefore, given an expansion $δ$ of hyperideals and a hyperideal reduction $ϕ$, a proper hyperideal $N$ of $\Re$ is called $ϕ$-$δ$-$r$-hyperideal if $a\cdot b\in N-ϕ(N)$ with $ann(a)=0$ implies that $b\inδ(N)$, for all $a,b\in\Re$. We investigate some of their properties and give some examples.

math.GM↗

arXiv2022

A UCB-based Tree Search Approach to Joint Verification-Correction Strategy for Large Scale Systems

Verification planning is a sequential decision-making problem that specifies a set of verification activities (VA) and correction activities (CA) at different phases of system development. While VAs are used to identify errors and defects, CAs also play important roles in system verification as they correct the identified errors and defects. However, current planning methods only consider VAs as decision choices. Because VAs and CAs have different activity spaces, planning a joint verification-correction strategy (JVCS) is still challenging, especially for large-size systems. Here we introduce a UCB-based tree search approach to search for near-optimal JVCSs. First, verification planning is simplified as repeatable bandit problems and an upper confidence bound rule for repeatable bandits (UCBRB) is presented with the optimal regret bound. Next, a tree search algorithm is proposed to search for feasible JVCSs. A tree-based ensemble learning model is also used to extend the tree search algorithm to handle local optimality issues. The proposed approach is evaluated on the notional case of a communication system.

cs.SE↗

arXiv2022

Value Function Spaces: Skill-Centric State Abstractions for Long-Horizon Reasoning

Reinforcement learning can train policies that effectively perform complex tasks. However for long-horizon tasks, the performance of these methods degrades with horizon, often necessitating reasoning over and chaining lower-level skills. Hierarchical reinforcement learning aims to enable this by providing a bank of low-level skills as action abstractions. Hierarchies can further improve on this by abstracting the space states as well. We posit that a suitable state abstraction should depend on the capabilities of the available lower-level policies. We propose Value Function Spaces: a simple approach that produces such a representation by using the value functions corresponding to each lower-level skill. These value functions capture the affordances of the scene, thus forming a representation that compactly abstracts task relevant information and robustly ignores distractors. Empirical evaluations for maze-solving and robotic manipulation tasks demonstrate that our approach improves long-horizon performance and enables better zero-shot generalization than alternative model-free and model-based methods.

cs.LG↗

arXiv2022

CI-AVSR: A Cantonese Audio-Visual Speech Dataset for In-car Command Recognition

With the rise of deep learning and intelligent vehicle, the smart assistant has become an essential in-car component to facilitate driving and provide extra functionalities. In-car smart assistants should be able to process general as well as car-related commands and perform corresponding actions, which eases driving and improves safety. However, there is a data scarcity issue for low resource languages, hindering the development of research and applications. In this paper, we introduce a new dataset, Cantonese In-car Audio-Visual Speech Recognition (CI-AVSR), for in-car command recognition in the Cantonese language with both video and audio data. It consists of 4,984 samples (8.3 hours) of 200 in-car commands recorded by 30 native Cantonese speakers. Furthermore, we augment our dataset using common in-car background noises to simulate real environments, producing a dataset 10 times larger than the collected one. We provide detailed statistics of both the clean and the augmented versions of our dataset. Moreover, we implement two multimodal baselines to demonstrate the validity of CI-AVSR. Experiment results show that leveraging the visual signal improves the overall performance of the model. Although our best model can achieve a considerable quality on the clean test set, the speech recognition quality on the noisy data is still inferior and remains as an extremely challenging task for real in-car speech recognition systems. The dataset and code will be released at https://github.com/HLTCHKUST/CI-AVSR.

cs.CL↗

arXiv2022

DeepChange: A Large Long-Term Person Re-Identification Benchmark with Clothes Change

Existing person re-identification (re-id) works mostly consider short-term application scenarios without clothes change. In real-world, however, we often dress differently across space and time. To solve this contrast, a few recent attempts have been made on long-term re-id with clothes change. Currently, one of the most significant limitations in this field is the lack of a large realistic benchmark. In this work, we contribute a large, realistic long-term person re-identification benchmark, named as DeepChange. It has several unique characteristics: (1) Realistic and rich personal appearance (e.g., clothes and hair style) and variations: Highly diverse clothes change and styles, with varying reappearing gaps in time from minutes to seasons, different weather conditions (e.g., sunny, cloudy, windy, rainy, snowy, extremely cold) and events (e.g., working, leisure, daily activities). (2) Rich camera setups: Raw videos were recorded by 17 outdoor varying resolution cameras operating in a real-world surveillance system. (3) The currently largest number of (17) cameras, (1, 121) identities, and (178, 407) bounding boxes, over the longest time span (12 months). Further, we investigate multimodal fusion strategies for tackling the clothes change challenge. Extensive experiments show that our fusion models outperform a wide variety of state-of-the-art models on DeepChange. Our dataset and documents are available at https://github.com/PengBoXiangShang/deepchange.

cs.CV↗

arXiv2022

Topological Micromotion of Floquet Quantum Systems

The Floquet Hamiltonian has often been used to describe a time-periodic system. Nevertheless, because the Floquet Hamiltonian depends on a micro-motion parameter, the Floquet Hamiltonian with a fixed micro-motion parameter cannot faithfully represent a driven system, which manifests as the anomalous edge states. Here we show that an accurate description of a Floquet system requires a set of Hamiltonian exhausting all values of the micro-motion parameter, and this micro-motion parameter can be viewed as an extra synthetic dimension of the system. Therefore, we show that a $d$-dimensional Floquet system can be described by a $d+1$-dimensional static Hamiltonian, and the advantage of this representation is that the periodic boundary condition is automatically imposed along the extra-dimension, which enables a straightforward definition of topological invariants. The topological invariant in the $d+1$-dimensional system can ensure a $d-1$-dimensional edge state of the $d$-dimensional Floquet system. Here we show two examples where the topological invariant is a three-dimensional Hopf invariant. We highlight that our scheme of classifying Floquet topology on the micro-motion space is different from the previous classification of Floquet topology on the time space.

quant-ph↗

arXiv2022

QA4QG: Using Question Answering to Constrain Multi-Hop Question Generation

Multi-hop question generation (MQG) aims to generate complex questions which require reasoning over multiple pieces of information of the input passage. Most existing work on MQG has focused on exploring graph-based networks to equip the traditional Sequence-to-sequence framework with reasoning ability. However, these models do not take full advantage of the constraint between questions and answers. Furthermore, studies on multi-hop question answering (QA) suggest that Transformers can replace the graph structure for multi-hop reasoning. Therefore, in this work, we propose a novel framework, QA4QG, a QA-augmented BART-based framework for MQG. It augments the standard BART model with an additional multi-hop QA module to further constrain the generated question. Our results on the HotpotQA dataset show that QA4QG outperforms all state-of-the-art models, with an increase of 8 BLEU-4 and 8 ROUGE points compared to the best results previously reported. Our work suggests the advantage of introducing pre-trained language models and QA module for the MQG task.

cs.CL↗

arXiv2022

DFT+$U$ within the framework of linear combination of numerical atomic orbitals

We present a formulation and implementation of the DFT+\textit{U} method within the framework of linear combination of numerical atomic orbitals (NAO). Our implementation not only enables single-point total energy and electronic-structure calculations but also provides access to atomic forces and stresses, hence allowing for full structure relaxations of periodic systems. Furthermore, our implementation allows one to deal with non-collinear spin texture, with the spin-orbit coupling (SOC) effect treated self-consistently. The key aspect behind our implementation is a suitable definition of the correlated subspace when multiple atomic orbitals with the same angular momentum are used, and this is addressed via the "Mulliken charge projector" constructed in terms of the first (most localized) atomic orbital within the $d/f$ angular momentum channel. The important Hubbard $U$ and Hund $J$ parameters can be estimated from a screened Coulomb potential of the Yukawa type, with the screening parameter either chosen semi-empirically or determined from the Thomas-Fermi screening model. Benchmark calculations are performed for four late transition metal monoxide bulk systems, i.e., MnO, FeO, CoO, and NiO, and for the 5$d$-electron compounds IrO$_2$. For the former type of systems, we check the performance of our DFT+$U$ implementation for calculating band gaps, magnetic moments, electronic band structures, as well as forces and stresses; for the latter, the efficacy of our DFT+$U$+SOC implementation is assessed. Systematic comparisons with available experimental results, and especially with the results from other implementation schemes are carried out, which demonstrate the validity of our NAO-based DFT+$U$ formalism and implementation.

cond-mat.mtrl-sci↗