Searcharxiv⌕ Search

arXiv subjects

Rui Yang

Publications and source records attributed to Rui Yang.

At least 253 records · Page 14Linked to original sources
arXiv2023

What is Essential for Unseen Goal Generalization of Offline Goal-conditioned RL?

Offline goal-conditioned RL (GCRL) offers a way to train general-purpose agents from fully offline datasets. In addition to being conservative within the dataset, the generalization ability to achieve unseen goals is another fundamental challenge for offline GCRL. However, to the best of our knowledge, this problem has not been well studied yet. In this paper, we study out-of-distribution (OOD) generalization of offline GCRL both theoretically and empirically to identify factors that are important. In a number of experiments, we observe that weighted imitation learning enjoys better generalization than pessimism-based offline RL method. Based on this insight, we derive a theory for OOD generalization, which characterizes several important design choices. We then propose a new offline GCRL method, Generalizable Offline goAl-condiTioned RL (GOAT), by combining the findings from our theoretical and empirical studies. On a new benchmark containing 9 independent identically distributed (IID) tasks and 17 OOD tasks, GOAT outperforms current state-of-the-art methods by a large margin.

cs.LG↗
arXiv2023

GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction

This paper aims to efficiently enable Large Language Models (LLMs) to use multimodal tools. Advanced proprietary LLMs, such as ChatGPT and GPT-4, have shown great potential for tool usage through sophisticated prompt engineering. Nevertheless, these models typically rely on prohibitive computational costs and publicly inaccessible data. To address these challenges, we propose the GPT4Tools based on self-instruct to enable open-source LLMs, such as LLaMA and OPT, to use tools. It generates an instruction-following dataset by prompting an advanced teacher with various multi-modal contexts. By using the Low-Rank Adaptation (LoRA) optimization, our approach facilitates the open-source LLMs to solve a range of visual problems, including visual comprehension and image generation. Moreover, we provide a benchmark to evaluate the ability of LLMs to use tools, which is performed in both zero-shot and fine-tuning ways. Extensive experiments demonstrate the effectiveness of our method on various language models, which not only significantly improves the accuracy of invoking seen tools, but also enables the zero-shot capacity for unseen tools. The code and demo are available at https://github.com/StevenGrove/GPT4Tools.

cs.CV↗
arXiv2023

Fidelity estimator, randomized benchmarking and ZNE for quantum pulses

Most previous research focused on designing pulse programs without considering the performance of individual elements or the final fidelity. To evaluate the performance of quantum pulses, it is required to know the noiseless results of the pulses. However, quantum pulses can implement unitary matrices that are not analytically known to the user, and pulse simulator usually comes with significant computational overhead. Consequently, determining fidelity of a pulse program is challenging without the knowledge of the ideal results. In this paper, we propose to use reversed pulses to evaluate the performance of quantum pulses, which can provide guidance to design pulse programs. By employing reversed pulses, we can ensure that, in the noiseless situation, the final quantum states are the same as the initial states. This method enables us to evaluate the fidelity of pulse programs by measuring the difference between the final states and the initial states. Such fidelity estimator can tell whether the results are meaningful for quantum pulses on real quantum machines. There are various quantum error correction (QEC) methods available for gate circuits; however, few studies have demonstrated QEC on pulse-level programs. In this paper, we use reversed pulses to implement zero noise extrapolation (ZNE) on pulse programs and demonstrate results for variational quantum eigensolver (VQE) tasks. The deviation from the idea energy value is reduced by an average of 54.1\% with our techniques.

quant-ph↗
arXiv2023

Analog quantum variational embedding classifier

Quantum machine learning has the potential to provide powerful algorithms for artificial intelligence. The pursuit of quantum advantage in quantum machine learning is an active area of research. For current noisy, intermediate-scale quantum (NISQ) computers, various quantum-classical hybrid algorithms have been proposed. One such previously proposed hybrid algorithm is a gate-based variational embedding classifier, which is composed of a classical neural network and a parameterized gate-based quantum circuit. We propose a quantum variational embedding classifier based on an analog quantum computer, where control signals vary continuously in time. In our algorithm, the classical data is transformed into the parameters of the time-varying Hamiltonian of the analog quantum computer by a linear transformation. The nonlinearity needed for a nonlinear classification problem is purely provided by the analog quantum computer through the nonlinear dependence of the final quantum state on the control parameters of the Hamiltonian. We performed numerical simulations that demonstrate the effectiveness of our algorithm for performing binary and multi-class classification on linearly inseparable datasets such as concentric circles and MNIST digits. Our classifier can reach accuracy comparable with the best classical classifiers. We find the performance of our classifier can be increased by increasing the number of qubits until the performance saturates and fluctuates. Moreover, the number of optimization parameters of our classifier scales linearly with the number of qubits. The increase of number of training parameters when the size increases is therefore not as fast as that of neural network. Our algorithm presents the possibility of using current quantum annealers for solving practical machine-learning problems, and it could also be useful to explore quantum advantage in quantum machine learning.

quant-ph↗
arXiv2023

ProtoDiv: Prototype-guided Division of Consistent Pseudo-bags for Whole-slide Image Classification

Due to the limitations of inadequate Whole-Slide Image (WSI) samples with weak labels, pseudo-bag-based multiple instance learning (MIL) appears as a vibrant prospect in WSI classification. However, the pseudo-bag dividing scheme, often crucial for classification performance, is still an open topic worth exploring. Therefore, this paper proposes a novel scheme, ProtoDiv, using a bag prototype to guide the division of WSI pseudo-bags. Rather than designing complex network architecture, this scheme takes a plugin-and-play approach to safely augment WSI data for effective training while preserving sample consistency. Furthermore, we specially devise an attention-based prototype that could be optimized dynamically in training to adapt to a classification task. We apply our ProtoDiv scheme on seven baseline models, and then carry out a group of comparison experiments on two public WSI datasets. Experiments confirm our ProtoDiv could usually bring obvious performance improvements to WSI classification.

cs.CV↗
arXiv2023

A Survey on Over-the-Air Computation

Communication and computation are often viewed as separate tasks. This approach is very effective from the perspective of engineering as isolated optimizations can be performed. However, for many computation-oriented applications, the main interest is a function of the local information at the devices, rather than the local information itself. In such scenarios, information theoretical results show that harnessing the interference in a multiple access channel for computation, i.e., over-the-air computation (OAC), can provide a significantly higher achievable computation rate than separating communication and computation tasks. Moreover, the gap between OAC and separation in terms of computation rate increases with more participating nodes. Given this motivation, in this study, we provide a comprehensive survey on practical OAC methods. After outlining fundamentals related to OAC, we discuss the available OAC schemes with their pros and cons. We provide an overview of the enabling mechanisms for achieving reliable computation in the wireless channel. Finally, we summarize the potential applications of OAC and point out some future directions.

cs.IT↗
arXiv2023

DSCA: A Dual-Stream Network with Cross-Attention on Whole-Slide Image Pyramids for Cancer Prognosis

The cancer prognosis on gigapixel Whole-Slide Images (WSIs) has always been a challenging task. To further enhance WSI visual representations, existing methods have explored image pyramids, instead of single-resolution images, in WSIs. In spite of this, they still face two major problems: high computational cost and the unnoticed semantical gap in multi-resolution feature fusion. To tackle these problems, this paper proposes to efficiently exploit WSI pyramids from a new perspective, the dual-stream network with cross-attention (DSCA). Our key idea is to utilize two sub-streams to process the WSI patches with two resolutions, where a square pooling is devised in a high-resolution stream to significantly reduce computational costs, and a cross-attention-based method is proposed to properly handle the fusion of dual-stream features. We validate our DSCA on three publicly-available datasets with a total number of 3,101 WSIs from 1,911 patients. Our experiments and ablation studies verify that (i) the proposed DSCA could outperform existing state-of-the-art methods in cancer prognosis, by an average C-Index improvement of around 4.6%; (ii) our DSCA network is more efficient in computation -- it has more learnable parameters (6.31M vs. 860.18K) but less computational costs (2.51G vs. 4.94G), compared to a typical existing multi-resolution network. (iii) the key components of DSCA, dual-stream and cross-attention, indeed contribute to our model's performance, gaining an average C-Index rise of around 2.0% while maintaining a relatively-small computational load. Our DSCA could serve as an alternative and effective tool for WSI-based cancer prognosis.

eess.IV↗
arXiv2023

MGA: Medical generalist agent through text-guided knowledge transformation

Multi-modal representation methods have achieved advanced performance in medical applications by extracting more robust features from multi-domain data. However, existing methods usually need to train additional branches for downstream tasks, which may increase the model complexities in clinical applications as well as introduce additional human inductive bias. Besides, very few studies exploit the rich clinical knowledge embedded in clinical daily reports. To this end, we propose a novel medical generalist agent, MGA, that can address three kinds of common clinical tasks via clinical reports knowledge transformation. Unlike the existing methods, MGA can easily adapt to different tasks without specific downstream branches when their corresponding annotations are missing. More importantly, we are the first attempt to use medical professional language guidance as a transmission medium to guide the agent's behavior. The proposed method is implemented on four well-known X-ray open-source datasets, MIMIC-CXR, CheXpert, MIMIC-CXR-JPG, and MIMIC-CXR-MS. Promising results are obtained, which validate the effectiveness of our proposed MGA. Code is available at: https://github.com/SZUHvern/MGA

cs.CV↗
arXiv2023

SSGD: A smartphone screen glass dataset for defect detection

Interactive devices with touch screen have become commonly used in various aspects of daily life, which raises the demand for high production quality of touch screen glass. While it is desirable to develop effective defect detection technologies to optimize the automatic touch screen production lines, the development of these technologies suffers from the lack of publicly available datasets. To address this issue, we in this paper propose a dedicated touch screen glass defect dataset which includes seven types of defects and consists of 2504 images captured in various scenarios.All data are captured with professional acquisition equipment on the fixed workstation. Additionally, we benchmark the CNN- and Transformer-based object detection frameworks on the proposed dataset to demonstrate the challenges of defect detection on high-resolution images. Dataset and related code will be available at https://github.com/Yangr116/SSGDataset.

cs.CV↗
arXiv2023

CoMesh: Fully-Decentralized Control for Sense-Trigger-Actuate Routines in Edge Meshes

While mesh networking for edge settings (e.g., smart buildings, farms, battlefields, etc.) has received much attention, the layer of control over such meshes remains largely centralized and cloud-based. This paper focuses on applications with sense-trigger-actuate (STA) workloads -- these are similar to the abstraction of routines popular in smart homes, but applied to larger-scale edge IoT deployments. We present CoMesh, which tackles the challenge of building local, non-cloud, and decentralized solutions for control of sense-trigger-actuate applications. At its core CoMesh uses an abstraction called k-groups to spread in a fine-grained way, the load of STA actions. Coordination within the k-group uses selective fast and cheap mechanisms rather than expensive off-the-shelf solutions. k-group selection is proactively dynamic, and occurs by using a combination of zero-message-exchange mechanisms (to reduce load) and locality sensitive hashing (to be aware of physical layout of devices). We analyze and theoretically prove the safety of CoMesh's mechanisms. Our evaluations using both simulation and Raspberry Pi lab deployments show that CoMesh is load-balanced, fast, and fault-tolerant.

cs.DC↗
arXiv2023

Robust Representation Learning by Clustering with Bisimulation Metrics for Visual Reinforcement Learning with Distractions

Recent work has shown that representation learning plays a critical role in sample-efficient reinforcement learning (RL) from pixels. Unfortunately, in real-world scenarios, representation learning is usually fragile to task-irrelevant distractions such as variations in background or viewpoint. To tackle this problem, we propose a novel clustering-based approach, namely Clustering with Bisimulation Metrics (CBM), which learns robust representations by grouping visual observations in the latent space. Specifically, CBM alternates between two steps: (1) grouping observations by measuring their bisimulation distances to the learned prototypes; (2) learning a set of prototypes according to the current cluster assignments. Computing cluster assignments with bisimulation metrics enables CBM to capture task-relevant information, as bisimulation metrics quantify the behavioral similarity between observations. Moreover, CBM encourages the consistency of representations within each group, which facilitates filtering out task-irrelevant information and thus induces robust representations against distractions. An appealing feature is that CBM can achieve sample-efficient representation learning even if multiple distractions exist simultaneously.Experiments demonstrate that CBM significantly improves the sample efficiency of popular visual RL algorithms and achieves state-of-the-art performance on both multiple and single distraction settings. The code is available at https://github.com/MIRALab-USTC/RL-CBM.

cs.LG↗
arXiv2023

Generalization in Visual Reinforcement Learning with the Reward Sequence Distribution

Generalization in partially observed markov decision processes (POMDPs) is critical for successful applications of visual reinforcement learning (VRL) in real scenarios. A widely used idea is to learn task-relevant representations that encode task-relevant information of common features in POMDPs, i.e., rewards and transition dynamics. As transition dynamics in the latent state space -- which are task-relevant and invariant to visual distractions -- are unknown to the agents, existing methods alternatively use transition dynamics in the observation space to extract task-relevant information in transition dynamics. However, such transition dynamics in the observation space involve task-irrelevant visual distractions, degrading the generalization performance of VRL methods. To tackle this problem, we propose the reward sequence distribution conditioned on the starting observation and the predefined subsequent action sequence (RSD-OA). The appealing features of RSD-OA include that: (1) RSD-OA is invariant to visual distractions, as it is conditioned on the predefined subsequent action sequence without task-irrelevant information from transition dynamics, and (2) the reward sequence captures long-term task-relevant information in both rewards and transition dynamics. Experiments demonstrate that our representation learning approach based on RSD-OA significantly improves the generalization performance on unseen environments, outperforming several state-of-the-arts on DeepMind Control tasks with visual distractions.

cs.LG↗
arXiv2023

Ice melting in salty water: layering and non-monotonic dependence on the mean salinity

The presence of salt in ocean water strongly affects the melt rate and the shape evolution of ice, both of utmost relevance in geophysical and ocean flow and thus for the climate. To get a better quantitative understanding of the physical mechanics at play in ice melting in salty water, we numerically investigate the lateral melting of an ice block in stably stratified saline water, using a realistic, nonlinear equation of state (EOS). The developing ice shape from our numerical results shows good agreement with the experiments and theory from Huppert & Turner (J. Fluid Mech. 100, 367 (1980)). Furthermore, we find that the melt rate of ice depends non-monotonically on the mean ambient salinity: It first decreases for increasing salt concentration until a local minimum is attained, and then increases again. This non-monotonic behavior of the ice melt rate is due to the competition among salinity-driven buoyancy, temperature-driven buoyancy, and salinity-induced stratification. We develop a theoretical model based on the energy balance which gives a prediction of the salt concentration for which the melt rate is minimal, and is consistent with our data. Our findings give insight into the interplay between phase transitions and double-diffusive convective flows.

physics.flu-dyn↗
arXiv2022

Learning-based Predictive Path Following Control for Nonlinear Systems Under Uncertain Disturbances

Accurate path following is challenging for autonomous robots operating in uncertain environments. Adaptive and predictive control strategies are crucial for a nonlinear robotic system to achieve high-performance path following control. In this paper, we propose a novel learning-based predictive control scheme that couples a high-level model predictive path following controller (MPFC) with a low-level learning-based feedback linearization controller (LB-FBLC) for nonlinear systems under uncertain disturbances. The low-level LB-FBLC utilizes Gaussian Processes to learn the uncertain environmental disturbances online and tracks the reference state accurately with a probabilistic stability guarantee. Meanwhile, the high-level MPFC exploits the linearized system model augmented with a virtual linear path dynamics model to optimize the evolution of path reference targets, and provides the reference states and controls for the low-level LB-FBLC. Simulation results illustrate the effectiveness of the proposed control strategy on a quadrotor path following task under unknown wind disturbances.

cs.RO↗
arXiv2022

RDP-Net: Region Detail Preserving Network for Change Detection

Change detection (CD) is an essential earth observation technique. It captures the dynamic information of land objects. With the rise of deep learning, convolutional neural networks (CNN) have shown great potential in CD. However, current CNN models introduce backbone architectures that lose detailed information during learning. Moreover, current CNN models are heavy in parameters, which prevents their deployment on edge devices such as UAVs. In this work, we tackle this issue by proposing RDP-Net: a region detail preserving network for CD. We propose an efficient training strategy that constructs the training tasks during the warmup period of CNN training and lets the CNN learn from easy to hard. The training strategy enables CNN to learn more powerful features with fewer FLOPs and achieve better performance. Next, we propose an effective edge loss that increases the penalty for errors on details and improves the network's attention to details such as boundary regions and small areas. Furthermore, we provide a CNN model with a brand new backbone that achieves the state-of-the-art empirical performance in CD with only 1.70M parameters. We hope our RDP-Net would benefit the practical CD applications on compact devices and could inspire more people to bring change detection to a new level with the efficient training strategy. The code and models are publicly available at https://github.com/Chnja/RDPNet.

eess.IV↗
arXiv2022

On Distribution Grid Optimal Power Flow Development and Integration

Due to changes in electric distribution grid operation, new operation regimes have been recommended. Distribution grid optimal power flow (DOPF) has received tremendous attention in the research community, yet it has not been fully adopted across the utility industry. Our paper recognizes this problem and suggests a development and integration procedure for DOPF. We propose development of DOPF as a three step procedure of 1) processing the grid, 2) obtaining a tractable solution, and 3) implementing multiple solution algorithms and benchmarking them to improve application reliability. For the integration of DOPF, we demonstrate how a DOPF federate may be developed that can be integrated in a co-simulation environment to mimic the real-world conditions and hence improve its practicality to be deployed in the field. To demonstrate the efficacy of the proposed methods, tests on IEEE 123-bus system are performed where the usage of tractable formulation in DOPF algorithm development and its comparison to the benchmark solution are demonstrated.

eess.SY↗
arXiv2022

Dual adaptive training of photonic neural networks

Photonic neural network (PNN) is a remarkable analog artificial intelligence (AI) accelerator that computes with photons instead of electrons to feature low latency, high energy efficiency, and high parallelism. However, the existing training approaches cannot address the extensive accumulation of systematic errors in large-scale PNNs, resulting in a significant decrease in model performance in physical systems. Here, we propose dual adaptive training (DAT) that allows the PNN model to adapt to substantial systematic errors and preserves its performance during the deployment. By introducing the systematic error prediction networks with task-similarity joint optimization, DAT achieves the high similarity mapping between the PNN numerical models and physical systems and high-accurate gradient calculations during the dual backpropagation training. We validated the effectiveness of DAT by using diffractive PNNs and interference-based PNNs on image classification tasks. DAT successfully trained large-scale PNNs under major systematic errors and preserved the model classification accuracies comparable to error-free systems. The results further demonstrated its superior performance over the state-of-the-art in situ training approaches. DAT provides critical support for constructing large-scale PNNs to achieve advanced architectures and can be generalized to other types of AI systems with analog computing errors.

cs.LG↗
arXiv2022

Nanomechanical Resonators: Toward Atomic Scale

The quest for realizing and manipulating ever smaller man-made movable structures and dynamical machines has spurred tremendous endeavors, led to important discoveries, and inspired researchers to venture to new grounds. Scientific feats and technological milestones of miniaturization of mechanical structures have been widely accomplished by advances in machining and sculpturing ever shrinking features out of bulk materials such as silicon. With the flourishing multidisciplinary field of low-dimensional nanomaterials, including one-dimensional (1D) nanowires/nanotubes, and two-dimensional (2D) atomic layers such as graphene/phosphorene, growing interests and sustained efforts have been devoted to creating mechanical devices toward the ultimate limit of miniaturization--genuinely down to the molecular or even atomic scale. These ultrasmall movable structures, particularly nanomechanical resonators that exploit the vibratory motion in these 1D and 2D nano-to-atomic-scale structures, offer exceptional device-level attributes, such as ultralow mass, ultrawide frequency tuning range, broad dynamic range, and ultralow power consumption, thus holding strong promises for both fundamental studies and engineering applications. In this Review, we offer a comprehensive overview and summary of this vibrant field, present the state-of-the-art devices and evaluate their specifications and performance, outline important achievements, and postulate future directions for studying these miniscule yet intriguing molecular-scale machines.

cond-mat.mes-hall↗