arXiv subjects

Tao Chen

Publications and source records attributed to Tao Chen.

At least 379 records · Page 21Linked to original sources

arXiv2023

$β$-DARTS++: Bi-level Regularization for Proxy-robust Differentiable Architecture Search

Neural Architecture Search has attracted increasing attention in recent years. Among them, differential NAS approaches such as DARTS, have gained popularity for the search efficiency. However, they still suffer from three main issues, that are, the weak stability due to the performance collapse, the poor generalization ability of the searched architectures, and the inferior robustness to different kinds of proxies. To solve the stability and generalization problems, a simple-but-effective regularization method, termed as Beta-Decay, is proposed to regularize the DARTS-based NAS searching process (i.e., $β$-DARTS). Specifically, Beta-Decay regularization can impose constraints to keep the value and variance of activated architecture parameters from being too large, thereby ensuring fair competition among architecture parameters and making the supernet less sensitive to the impact of input on the operation set. In-depth theoretical analyses on how it works and why it works are provided. Comprehensive experiments validate that Beta-Decay regularization can help to stabilize the searching process and makes the searched network more transferable across different datasets. To address the robustness problem, we first benchmark different NAS methods under a wide range of proxy data, proxy channels, proxy layers and proxy epochs, since the robustness of NAS under different kinds of proxies has not been explored before. We then conclude some interesting findings and find that $β$-DARTS always achieves the best result among all compared NAS methods under almost all proxies. We further introduce the novel flooding regularization to the weight optimization of $β$-DARTS (i.e., Bi-level regularization), and experimentally and theoretically verify its effectiveness for improving the proxy robustness of differentiable NAS.

cs.LG↗

arXiv2023

Optimizing nonadiabatic geometric quantum gates against off-resonance error by dynamical correction in a silicon-based spin qubit

Geometric quantum gates are performed by using the geometric phase, making them particularly robust to the pulse amplitude error due to the intrinsic global property. However, in many systems, such as the silicon-based spin qubits, the off-resonance error is the dominant noise, which can cause dephasing and is always difficult to deal with for a geometric gate. Thus how to deal with the off-resonance error is very significant for the application of the geometric gates. A recent work in \emph{Phy. Rev. Appl. 16, 044005 (2021)} reveals that by inserting two $π$-pulse dynamically corrected sequences into the evolution paths, the holonomic quantum gate is effective to suppress the pulse amplitude error, however it is still useless for combating the off-resonance error. Inspired by this work, we combine using the techniques of dynamical correction and path design. Surprisingly, we find that by picking up a specific evolution path inserted by only a $π$-pulse dynamically corrected sequence, the obtained optimized geometric gate is robust to the off-resonance error, assuming the noise is static. Further, by calculating the filter function considering the realistic $1/f$-type noise in silicon, the related results show that the performance of the optimized geometric gate can also surpass both the conventional geometric gate and the naive dynamical gate constructed without using the geometric phase. Our results indicate dynamical correction is an powerful tool to improve the geometric gate.

quant-ph↗

arXiv2023

Do Performance Aspirations Matter for Guiding Software Configuration Tuning?

Configurable software systems can be tuned for better performance. Leveraging on some Pareto optimizers, recent work has shifted from tuning for a single, time-related performance objective to two intrinsically different objectives that assess distinct performance aspects of the system, each with varying aspirations. Before we design better optimizers, a crucial engineering decision to make therein is how to handle the performance requirements with clear aspirations in the tuning process. For this, the community takes two alternative optimization models: either quantifying and incorporating the aspirations into the search objectives that guide the tuning, or not considering the aspirations during the search but purely using them in the later decision-making process only. However, despite being a crucial decision that determines how an optimizer can be designed and tailored, there is a rather limited understanding of which optimization model should be chosen under what particular circumstance, and why. In this paper, we seek to close this gap. Firstly, we do that through a review of over 426 papers in the literature and 14 real-world requirements datasets. Drawing on these, we then conduct a comprehensive empirical study that covers 15 combinations of the state-of-the-art performance requirement patterns, four types of aspiration space, three Pareto optimizers, and eight real-world systems/environments, leading to 1,296 cases of investigation. We found that (1) the realism of aspirations is the key factor that determines whether they should be used to guide the tuning; (2) the given patterns and the position of the realistic aspirations in the objective landscape are less important for the choice, but they do matter to the extents of improvement; (3) the available tuning budget can also influence the choice for unrealistic aspirations but it is insignificant under realistic ones.

cs.SE↗

arXiv2023

End-to-End 3D Dense Captioning with Vote2Cap-DETR

3D dense captioning aims to generate multiple captions localized with their associated object regions. Existing methods follow a sophisticated ``detect-then-describe'' pipeline equipped with numerous hand-crafted components. However, these hand-crafted components would yield suboptimal performance given cluttered object spatial and class distributions among different scenes. In this paper, we propose a simple-yet-effective transformer framework Vote2Cap-DETR based on recent popular \textbf{DE}tection \textbf{TR}ansformer (DETR). Compared with prior arts, our framework has several appealing advantages: 1) Without resorting to numerous hand-crafted components, our method is based on a full transformer encoder-decoder architecture with a learnable vote query driven object decoder, and a caption decoder that produces the dense captions in a set-prediction manner. 2) In contrast to the two-stage scheme, our method can perform detection and captioning in one-stage. 3) Without bells and whistles, extensive experiments on two commonly used datasets, ScanRefer and Nr3D, demonstrate that our Vote2Cap-DETR surpasses current state-of-the-arts by 11.13\% and 7.11\% in CIDEr@0.5IoU, respectively. Codes will be released soon.

cs.CV↗

arXiv2022

Stochastic differential variational inequalities with applications

In this paper, we introduce and study a stochastic differential variational inequality (SDVI) which consists of a stochastic differential equation and a stochastic variational inequality. We obtain the existence and uniqueness of the solutions for SDVI by using the iteration method and Gronwall's inequality. Moreover, we show the convergence of Euler scheme for solving SDVI under some mild conditions. Finally, we apply the obtained results to solve the electrical circuits with diodes and the collapse of the bridge problems in stochastic environment.

math.OC↗

arXiv2022

Towards Accurate Ground Plane Normal Estimation from Ego-Motion

In this paper, we introduce a novel approach for ground plane normal estimation of wheeled vehicles. In practice, the ground plane is dynamically changed due to braking and unstable road surface. As a result, the vehicle pose, especially the pitch angle, is oscillating from subtle to obvious. Thus, estimating ground plane normal is meaningful since it can be encoded to improve the robustness of various autonomous driving tasks (e.g., 3D object detection, road surface reconstruction, and trajectory planning). Our proposed method only uses odometry as input and estimates accurate ground plane normal vectors in real time. Particularly, it fully utilizes the underlying connection between the ego pose odometry (ego-motion) and its nearby ground plane. Built on that, an Invariant Extended Kalman Filter (IEKF) is designed to estimate the normal vector in the sensor's coordinate. Thus, our proposed method is simple yet efficient and supports both camera- and inertial-based odometry algorithms. Its usability and the marked improvement of robustness are validated through multiple experiments on public datasets. For instance, we achieve state-of-the-art accuracy on KITTI dataset with the estimated vector error of 0.39°. Our code is available at github.com/manymuch/ground_normal_filter.

cs.CV↗

arXiv2022

Boundary vorticity dynamics of two-phase viscous flow

From the Navier-Stokes-Korteweg (NSK) equations, the exact relations between the fundamental surface physical quantities for two-phase viscous flow with diffuse interface are derived, including density gradient, shear stress, vorticity, pressure, enstrophy flux and surface curvature. These theoretical results provide a solid foundation of the boundary/interfacial vorticity dynamics and a new tool for analysis of complex interfacial phenomena in two-phase viscous flows. To demonstrate the application of the developed results, simulation of a droplet impacting and spreading on a solid wall is conducted by using a recently developed well-balanced discrete unified gas kinetic scheme (WB-DUGKS), focusing on spreading process when the separation bubbles form inside the droplet. The distributions of shear stress, pressure and enstrophy flux at the interface and wall are analyzed, particularly near the moving contact points and other characteristic points. This example gives an unique perspective to the physics of droplet impingement on a wall.

physics.flu-dyn↗

arXiv2022

Instance-aware Model Ensemble With Distillation For Unsupervised Domain Adaptation

The linear ensemble based strategy, i.e., averaging ensemble, has been proposed to improve the performance in unsupervised domain adaptation tasks. However, a typical UDA task is usually challenged by dynamically changing factors, such as variable weather, views, and background in the unlabeled target domain. Most previous ensemble strategies ignore UDA's dynamic and uncontrollable challenge, facing limited feature representations and performance bottlenecks. To enhance the model, adaptability between domains and reduce the computational cost when deploying the ensemble model, we propose a novel framework, namely Instance aware Model Ensemble With Distillation, IMED, which fuses multiple UDA component models adaptively according to different instances and distills these components into a small model. The core idea of IMED is a dynamic instance aware ensemble strategy, where for each instance, a nonlinear fusion subnetwork is learned that fuses the extracted features and predicted labels of multiple component models. The nonlinear fusion method can help the ensemble model handle dynamically changing factors. After learning a large capacity ensemble model with good adaptability to different changing factors, we leverage the ensemble teacher model to guide the learning of a compact student model by knowledge distillation. Furthermore, we provide the theoretical analysis of the validity of IMED for UDA. Extensive experiments conducted on various UDA benchmark datasets, e.g., Office 31, Office Home, and VisDA 2017, show the superiority of the model based on IMED to the state of the art methods under the comparable computation cost.

cs.CV↗

arXiv2022

Graphene-Induced Energy Transfer for Quantitative Membrane Biophysics at Sub-Nanometer Resolution

Graphene-induced energy transfer (GIET) is a recently developed fluorescence-spectroscopic technique that achieves sub-nanometric optical localization of fluorophores along the optical axis of a microscope. GIET is based on the near-field energy transfer from an optically excited fluorescent molecule to a single sheet of graphene. It has been successfully used for estimating inter-leaflet distances of single lipid bilayers, and for investigating the membrane organization of living mitochondria. In this study, we use GIET to measure the cholesterol-induced subtle changes of membrane thickness at the nanoscale. We quantify membrane thickness variations in supported lipid bilayers (SLBs) as a function of lipid composition and increasing cholesterol content. Our findings demonstrate that GIET is an extremely sensitive tool for investigating nanometric structural changes in bio-membranes.

physics.optics↗

arXiv2022

Cross-Subject Emotion Recognition with Sparsely-Labeled Peripheral Physiological Data Using SHAP-Explained Tree Ensembles

There are still many challenges of emotion recognition using physiological data despite the substantial progress made recently. In this paper, we attempted to address two major challenges. First, in order to deal with the sparsely-labeled physiological data, we first decomposed the raw physiological data using signal spectrum analysis, based on which we extracted both complexity and energy features. Such a procedure helped reduce noise and improve feature extraction effectiveness. Second, in order to improve the explainability of the machine learning models in emotion recognition with physiological data, we proposed Light Gradient Boosting Machine (LightGBM) and SHapley Additive exPlanations (SHAP) for emotion prediction and model explanation, respectively. The LightGBM model outperformed the eXtreme Gradient Boosting (XGBoost) model on the public Database for Emotion Analysis using Physiological signals (DEAP) with f1-scores of 0.814, 0.823, and 0.860 for binary classification of valence, arousal, and liking, respectively, with cross-subject validation using eight peripheral physiological signals. Furthermore, the SHAP model was able to identify the most important features in emotion recognition, and revealed the relationships between the predictor variables and the response variables in terms of their main effects and interaction effects. Therefore, the results of the proposed model not only had good performance using peripheral physiological data, but also gave more insights into the underlying mechanisms in recognizing emotions.

cs.HC↗

arXiv2022

Pre-Trained Language Models for Interactive Decision-Making

Language model (LM) pre-training is useful in many language processing tasks. But can pre-trained LMs be further leveraged for more general machine learning problems? We propose an approach for using LMs to scaffold learning and generalization in general sequential decision-making problems. In this approach, goals and observations are represented as a sequence of embeddings, and a policy network initialized with a pre-trained LM predicts the next action. We demonstrate that this framework enables effective combinatorial generalization across different environments and supervisory modalities. We begin by assuming access to a set of expert demonstrations, and show that initializing policies with LMs and fine-tuning them via behavior cloning improves task completion rates by 43.6% in the VirtualHome environment. Next, we integrate an active data gathering procedure in which agents iteratively interact with the environment, relabel past "failed" experiences with new goals, and update their policies in a self-supervised loop. Active data gathering further improves combinatorial generalization, outperforming the best baseline by 25.1%. Finally, we explain these results by investigating three possible factors underlying the effectiveness of the LM-based policy. We find that sequential input representations (vs. fixed-dimensional feature vectors) and LM-based weight initialization are both important for generalization. Surprisingly, however, the format of the policy inputs encoding (e.g. as a natural language string vs. an arbitrary sequential encoding) has little influence. Together, these results suggest that language modeling induces representations that are useful for modeling not just language, but also goals and plans; these representations can aid learning and generalization even outside of language processing.

cs.LG↗

arXiv2022

Coordinates Are NOT Lonely -- Codebook Prior Helps Implicit Neural 3D Representations

Implicit neural 3D representation has achieved impressive results in surface or scene reconstruction and novel view synthesis, which typically uses the coordinate-based multi-layer perceptrons (MLPs) to learn a continuous scene representation. However, existing approaches, such as Neural Radiance Field (NeRF) and its variants, usually require dense input views (i.e. 50-150) to obtain decent results. To relive the over-dependence on massive calibrated images and enrich the coordinate-based feature representation, we explore injecting the prior information into the coordinate-based network and introduce a novel coordinate-based model, CoCo-INR, for implicit neural 3D representation. The cores of our method are two attention modules: codebook attention and coordinate attention. The former extracts the useful prototypes containing rich geometry and appearance information from the prior codebook, and the latter propagates such prior information into each coordinate and enriches its feature representation for a scene or object surface. With the help of the prior information, our method can render 3D views with more photo-realistic appearance and geometries than the current methods using fewer calibrated images available. Experiments on various scene reconstruction datasets, including DTU and BlendedMVS, and the full 3D head reconstruction dataset, H3DS, demonstrate the robustness under fewer input views and fine detail-preserving capability of our proposed method.

cs.CV↗

arXiv2022

Retrieval Augmentation for T5 Re-ranker using External Sources

Retrieval augmentation has shown promising improvements in different tasks. However, whether such augmentation can assist a large language model based re-ranker remains unclear. We investigate how to augment T5-based re-rankers using high-quality information retrieved from two external corpora -- a commercial web search engine and Wikipedia. We empirically demonstrate how retrieval augmentation can substantially improve the effectiveness of T5-based re-rankers for both in-domain and zero-shot out-of-domain re-ranking tasks.

cs.IR↗

arXiv2022

Stimulative Training of Residual Networks: A Social Psychology Perspective of Loafing

Residual networks have shown great success and become indispensable in today's deep models. In this work, we aim to re-investigate the training process of residual networks from a novel social psychology perspective of loafing, and further propose a new training strategy to strengthen the performance of residual networks. As residual networks can be viewed as ensembles of relatively shallow networks (i.e., \textit{unraveled view}) in prior works, we also start from such view and consider that the final performance of a residual network is co-determined by a group of sub-networks. Inspired by the social loafing problem of social psychology, we find that residual networks invariably suffer from similar problem, where sub-networks in a residual network are prone to exert less effort when working as part of the group compared to working alone. We define this previously overlooked problem as \textit{network loafing}. As social loafing will ultimately cause the low individual productivity and the reduced overall performance, network loafing will also hinder the performance of a given residual network and its sub-networks. Referring to the solutions of social psychology, we propose \textit{stimulative training}, which randomly samples a residual sub-network and calculates the KL-divergence loss between the sampled sub-network and the given residual network, to act as extra supervision for sub-networks and make the overall goal consistent. Comprehensive empirical results and theoretical analyses verify that stimulative training can well handle the loafing problem, and improve the performance of a residual network by improving the performance of its sub-networks. The code is available at https://github.com/Sunshine-Ye/NIPS22-ST .

cs.CV↗

arXiv2022

Experimental Implementation of Noncyclic and Nonadiabatic Geometric Quantum Gates in a Superconducting Circuit

Quantum gates based on geometric phases possess intrinsic noise-resilience features and therefore attract much attention. However, the implementations of previous geometric quantum computation typically require a long pulse time of gates. As a result, their experimental control inevitably suffers from the cumulative disturbances of systematic errors due to excessive time consumption. Here, we experimentally implement a set of noncyclic and nonadiabatic geometric quantum gates in a superconducting circuit, which greatly shortens the gate time. And also, we experimentally verify that our universal single-qubit geometric gates are more robust to both the Rabi frequency error and qubit frequency shift-induced error, compared to the conventional dynamical gates, by using the randomized benchmarking method. Moreover, this scheme can be utilized to construct two-qubit geometric operations, while the generation of the maximally entangled Bell states is demonstrated. Therefore, our results provide a promising routine to achieve fast, high-fidelity, and error-resilient quantum gates in superconducting quantum circuits.

quant-ph↗

arXiv2022

Distilling Task-specific Logical Rules from Large Pre-trained Models

Logical rules, both transferable and explainable, are widely used as weakly supervised signals for many downstream tasks such as named entity tagging. To reduce the human effort of writing rules, previous researchers adopt an iterative approach to automatically learn logical rules from several seed rules. However, obtaining more seed rules can only be accomplished by extra human annotation with heavy costs. Limited by the size and quality of the seed rules, the model performance of previous systems is bounded. In this paper, we develop a novel framework STREAM to distill task-specific logical rules from large pre-trained models. Specifically, we borrow recent prompt-based language models as the knowledge expert to yield initial seed rules, and based on the formed high-quality instance pool that acts as an intermediary role, we keep teaching the expert to fit our task and learning task-specific logical rules. Experiments on three public named entity tagging benchmarks demonstrate the effectiveness of our proposed framework. With several predefined prompt templates, our system has gained significant improvements over previous state-of-the-art methods.

cs.CL↗

arXiv2022

Multifaceted Hierarchical Report Identification for Non-Functional Bugs in Deep Learning Frameworks

Non-functional bugs (e.g., performance- or accuracy-related bugs) in Deep Learning (DL) frameworks can lead to some of the most devastating consequences. Reporting those bugs on a repository such as GitHub is a standard route to fix them. Yet, given the growing number of new GitHub reports for DL frameworks, it is intrinsically difficult for developers to distinguish those that reveal non-functional bugs among the others, and assign them to the right contributor for investigation in a timely manner. In this paper, we propose MHNurf - an end-to-end tool for automatically identifying non-functional bug related reports in DL frameworks. The core of MHNurf is a Multifaceted Hierarchical Attention Network (MHAN) that tackles three unaddressed challenges: (1) learning the semantic knowledge, but doing so by (2) considering the hierarchy (e.g., words/tokens in sentences/statements) and focusing on the important parts (i.e., words, tokens, sentences, and statements) of a GitHub report, while (3) independently extracting information from different types of features, i.e., content, comment, code, command, and label. To evaluate MHNurf, we leverage 3,721 GitHub reports from five DL frameworks for conducting experiments. The results show that MHNurf works the best with a combination of content, comment, and code, which considerably outperforms the classic HAN where only the content is used. MHNurf also produces significantly more accurate results than nine other state-of-the-art classifiers with strong statistical significance, i.e., up to 71% AUC improvement and has the best Scott-Knott rank on four frameworks while 2nd on the remaining one. To facilitate reproduction and promote future research, we have made our dataset, code, and detailed supplementary results publicly available at: https://github.com/ideas-labo/APSEC2022-MHNurf.

cs.SE↗

arXiv2022

Data-Driven Nonparametric Robust Control under Dependence Uncertainty

We consider a multi-period stochastic control problem where the multivariate driving stochastic factor of the system has known marginal distributions but uncertain dependence structure. To solve the problem, we propose to implement the nonparametric adaptive robust control framework. We aim to find the optimal control against the worst-case copulae in a sequence of shrinking uncertainty sets which are generated from continuously observing the data. Then, we use a stochastic gradient descent ascent algorithm to numerically handle the corresponding high dimensional dynamic inf-sup optimization problem. We present the numerical results in the context of utility maximization and show that the controller benefits from knowing more information about the uncertain model.

math.OC↗