arXiv subjects

Chenyang Lu

Publications and source records attributed to Chenyang Lu.

At least 37 records · Page 2Linked to original sources

arXiv2023

Content-aware Token Sharing for Efficient Semantic Segmentation with Vision Transformers

This paper introduces Content-aware Token Sharing (CTS), a token reduction approach that improves the computational efficiency of semantic segmentation networks that use Vision Transformers (ViTs). Existing works have proposed token reduction approaches to improve the efficiency of ViT-based image classification networks, but these methods are not directly applicable to semantic segmentation, which we address in this work. We observe that, for semantic segmentation, multiple image patches can share a token if they contain the same semantic class, as they contain redundant information. Our approach leverages this by employing an efficient, class-agnostic policy network that predicts if image patches contain the same semantic class, and lets them share a token if they do. With experiments, we explore the critical design choices of CTS and show its effectiveness on the ADE20K, Pascal Context and Cityscapes datasets, various ViT backbones, and different segmentation decoders. With Content-aware Token Sharing, we are able to reduce the number of processed tokens by up to 44%, without diminishing the segmentation quality.

cs.CV↗

arXiv2023

Universal enhancement of vacancy diffusion by Mn inducing anomalous Friedel oscillation in concentrated solid-solution alloys

We present a proof-of-principle demonstration of a universal law for the element Mn, which greatly enhances vacancy diffusion through an anomalous Friedel Oscillation effect in a series of Ni-based concentrated solid-solution alloys, regardless of the type of atom involved. The antiferromagnetic element Mn possesses a unique half-filled 3d electron structure, creating split virtual bound states near the Fermi energy level and producing a large local magnetic moment after vacancy formation. The resultant electron spin oscillations reduce the number of electrons involved in charge density oscillations, destroying charge screening and lowering potential interaction at the saddle point between the vacancy and diffusing atom. This ultimately facilitates vacancy diffusion by reducing energy level variations of conduction band electrons during the diffusion process. These findings offer valuable insights into atom diffusion mechanisms and open up new avenues for manipulating defect properties through unique element design, thereby enabling the creation of high-performance alloys in a broad range of fields.

cond-mat.mtrl-sci↗

arXiv2022

Self-explaining Hierarchical Model for Intraoperative Time Series

Major postoperative complications are devastating to surgical patients. Some of these complications are potentially preventable via early predictions based on intraoperative data. However, intraoperative data comprise long and fine-grained multivariate time series, prohibiting the effective learning of accurate models. The large gaps associated with clinical events and protocols are usually ignored. Moreover, deep models generally lack transparency. Nevertheless, the interpretability is crucial to assist clinicians in planning for and delivering postoperative care and timely interventions. Towards this end, we propose a hierarchical model combining the strength of both attention and recurrent models for intraoperative time series. We further develop an explanation module for the hierarchical model to interpret the predictions by providing contributions of intraoperative data in a fine-grained manner. Experiments on a large dataset of 111,888 surgeries with multiple outcomes and an external high-resolution ICU dataset show that our model can achieve strong predictive performance (i.e., high accuracy) and offer robust interpretations (i.e., high transparency) for predicted outcomes based on intraoperative time series.

cs.LG↗

arXiv2022

Predicting Intraoperative Hypoxemia with Hybrid Inference Sequence Autoencoder Networks

We present an end-to-end model using streaming physiological time series to predict near-term risk for hypoxemia, a rare, but life-threatening condition known to cause serious patient harm during surgery. Inspired by the fact that a hypoxemia event is defined based on a future sequence of low SpO2 (i.e., blood oxygen saturation) instances, we propose the hybrid inference network (hiNet) that makes hybrid inference on both future low SpO2 instances and hypoxemia outcomes. hiNet integrates 1) a joint sequence autoencoder that simultaneously optimizes a discriminative decoder for label prediction, and 2) two auxiliary decoders trained for data reconstruction and forecast, which seamlessly learn contextual latent representations that capture the transition from present states to future states. All decoders share a memory-based encoder that helps capture the global dynamics of patient measurement. For a large surgical cohort of 72,081 surgeries at a major academic medical center, our model outperforms strong baselines including the model used by the state-of-the-art hypoxemia prediction system. With its capability to make real-time predictions of near-term hypoxemic at clinically acceptable alarm rates, hiNet shows promise in improving clinical decision making and easing burden of perioperative care.

cs.LG↗

arXiv2022

Adaptive Edge Offloading for Image Classification Under Rate Limit

This paper considers a setting where embedded devices are used to acquire and classify images. Because of limited computing capacity, embedded devices rely on a parsimonious classification model with uneven accuracy. When local classification is deemed inaccurate, devices can decide to offload the image to an edge server with a more accurate but resource-intensive model. Resource constraints, e.g., network bandwidth, however, require regulating such transmissions to avoid congestion and high latency. The paper investigates this offloading problem when transmissions regulation is through a token bucket, a mechanism commonly used for such purposes. The goal is to devise a lightweight, online offloading policy that optimizes an application-specific metric (e.g., classification accuracy) under the constraints of the token bucket. The paper develops a policy based on a Deep Q-Network (DQN), and demonstrates both its efficacy and the feasibility of its deployment on embedded devices. Of note is the fact that the policy can handle complex input patterns, including correlation in image arrivals and classification accuracy. The evaluation is carried out by performing image classification over a local testbed using synthetic traces generated from the ImageNet image classification benchmark. Implementation of this work is available at https://github.com/qiujiaming315/edgeml-dqn.

cs.DC↗

arXiv2022

HiPAL: A Deep Framework for Physician Burnout Prediction Using Activity Logs in Electronic Health Records

Burnout is a significant public health concern affecting nearly half of the healthcare workforce. This paper presents the first end-to-end deep learning framework for predicting physician burnout based on electronic health record (EHR) activity logs, digital traces of physician work activities that are available in any EHR system. In contrast to prior approaches that exclusively relied on surveys for burnout measurement, our framework directly learns deep representations of physician behaviors from large-scale clinician activity logs to predict burnout. We propose the Hierarchical burnout Prediction based on Activity Logs (HiPAL), featuring a pre-trained time-dependent activity embedding mechanism tailored for activity logs and a hierarchical predictive model, which mirrors the natural hierarchical structure of clinician activity logs and captures physicians' evolving burnout risk at both short-term and long-term levels. To utilize the large amount of unlabeled activity logs, we propose a semi-supervised framework that learns to transfer knowledge extracted from unlabeled clinician activities to the HiPAL-based prediction model. The experiment on over 15 million clinician activity logs collected from the EHR at a large academic medical center demonstrates the advantages of our proposed framework in predictive performance of physician burnout and training efficiency over state-of-the-art approaches.

cs.LG↗

arXiv2022

Self-Supervised Road Layout Parsing with Graph Auto-Encoding

Aiming for higher-level scene understanding, this work presents a neural network approach that takes a road-layout map in bird's-eye-view as input, and predicts a human-interpretable graph that represents the road's topological layout. Our approach elevates the understanding of road layouts from pixel level to the level of graphs. To achieve this goal, an image-graph-image auto-encoder is utilized. The network is designed to learn to regress the graph representation at its auto-encoder bottleneck. This learning is self-supervised by an image reconstruction loss, without needing any external manual annotations. We create a synthetic dataset containing common road layout patterns and use it for training of the auto-encoder in addition to the real-world Argoverse dataset. By using this additional synthetic dataset, which conceptually captures human knowledge of road layouts and makes this available to the network for training, we are able to stabilize and further improve the performance of topological road layout understanding on the real-world Argoverse dataset. The evaluation shows that our approach exhibits comparable performance to a strong fully-supervised baseline.

cs.CV↗

arXiv2021

Tuned Vacancy Diffusion by Mn via Anomalous Friedel Oscillations in NiCoFeCrMn High Entropy Alloys

Vacancy is known to play a critical role in solid state diffusion and determine many properties of the materials. We report a proof-of-principle result on Mn in the role of tuning vacancy diffusion by an anomalous Friedel Oscillations effect in an equivalent NiCoFeCrMn high entropy alloy. The vacancy induced Friedel Oscillations is destroyed by the presence of the first nearest neighboring Mn, opening a channel for long-range influence on the surrounding atoms from the vacancy, resulting in a lower vacancy diffusion barrier energy, despite of the atom type. The effects lead to a remarkable diffusion overlaps of the two type defects, vacancy and interstitial, enhancing their recombination, which can explain well for the experimental results with suppression of vacancy clusters under irradiation. This finding presents new understanding on the atomic diffusion mechanism and thus suggests novel concepts for tuning defect properties in the design of advanced materials with outstanding properties, in a broad field.

cond-mat.mtrl-sci↗

arXiv2021

Part-aware Panoptic Segmentation

In this work, we introduce the new scene understanding task of Part-aware Panoptic Segmentation (PPS), which aims to understand a scene at multiple levels of abstraction, and unifies the tasks of scene parsing and part parsing. For this novel task, we provide consistent annotations on two commonly used datasets: Cityscapes and Pascal VOC. Moreover, we present a single metric to evaluate PPS, called Part-aware Panoptic Quality (PartPQ). For this new task, using the metric and annotations, we set multiple baselines by merging results of existing state-of-the-art methods for panoptic segmentation and part segmentation. Finally, we conduct several experiments that evaluate the importance of the different levels of abstraction in this single task.

cs.CV↗

arXiv2021

Radiation-tolerant high-entropy alloys via interstitial-solute-induced chemical heterogeneities

High-entropy alloys (HEAs) composed of multiple principal elements have been shown to offer improved radiation resistance over their elemental or dilute-solution counterparts. Using NiCoFeCrMn HEA as a model, here we introduce carbon and nitrogen interstitial alloying elements to impart chemical heterogeneities in the form of the local chemical order (LCO) and associated compositional variations. Density functional theory simulations predict chemical short-range order (CSRO) (nearest neighbors and the next couple of atomic shells) surrounding C and N, due to the chemical affinity of C with (Co, Fe) and N with (Cr, Mn). Atomic-resolution chemical mapping of the elemental distribution confirms marked compositional variations well beyond statistical fluctuations. Ni+ irradiation experiments at elevated temperatures demonstrate a remarkable reduction in void swelling by at least one order of magnitude compared to the base HEA without C and N alloying. The underlying mechanism is that the interstitial-solute-induced chemical heterogeneities roughen the lattice as well as the energy landscape, impeding the movements of, and constraining the path lanes for, the normally fast-moving self-interstitials and their clusters. The irradiation-produced interstitials and vacancies therefore recombine more readily, delaying void formation. Our findings thus open a promising avenue towards highly radiation-tolerant alloys.

cond-mat.mtrl-sci↗

arXiv2021

Impact of Distributed Rate Limiting on Load Distribution in a Latency-sensitive Messaging Service

The cloud's flexibility and promise of seamless auto-scaling notwithstanding, its ability to meet service level objectives (SLOs) typically calls for some form of control in resource usage. This seemingly traditional problem gives rise to new challenges in a cloud setting, and in particular a subtle yet significant trade-off involving load-distribution decisions (the distribution of workload across available cloud resources to optimize performance), and rate limiting (the capping of individual workloads to prevent global over-commitment). This paper investigates that trade-off through the design and implementation of a real-time messaging system motivated by Internet-of-Things (IoT) applications, and demonstrates a solution capable of realizing an effective compromise. The paper's contributions are in both explicating the source of this trade-off, and in demonstrating a possible solution.

cs.NI↗

arXiv2020

Image-Graph-Image Translation via Auto-Encoding

This work presents the first convolutional neural network that learns an image-to-graph translation task without needing external supervision. Obtaining graph representations of image content, where objects are represented as nodes and their relationships as edges, is an important task in scene understanding. Current approaches follow a fully-supervised approach thereby requiring meticulous annotations. To overcome this, we are the first to present a self-supervised approach based on a fully-differentiable auto-encoder in which the bottleneck encodes the graph's nodes and edges. This self-supervised approach can currently encode simple line drawings into graphs and obtains comparable results to a fully-supervised baseline in terms of F1 score on triplet matching. Besides these promising results, we provide several directions for future research on how our approach can be extended to cover more complex imagery.

cs.CV↗

arXiv2020

Real-Time Edge Classification: Optimal Offloading under Token Bucket Constraints

To deploy machine learning-based algorithms for real-time applications with strict latency constraints, we consider an edge-computing setting where a subset of inputs are offloaded to the edge for processing by an accurate but resource-intensive model, and the rest are processed only by a less-accurate model on the device itself. Both models have computational costs that match available compute resources, and process inputs with low-latency. But offloading incurs network delays, and to manage these delays to meet application deadlines, we use a token bucket to constrain the average rate and burst length of transmissions from the device. We introduce a Markov Decision Process-based framework to make offload decisions under these constraints, based on the local model's confidence and the token bucket state, with the goal of minimizing a specified error measure for the application. Beyond isolated decisions for individual devices, we also propose approaches to allow multiple devices connected to the same access switch to share their bursting allocation. We evaluate and analyze the policies derived using our framework on the standard ImageNet image classification benchmark.

cs.LG↗

arXiv2020

Cityscapes-Panoptic-Parts and PASCAL-Panoptic-Parts datasets for Scene Understanding

In this technical report, we present two novel datasets for image scene understanding. Both datasets have annotations compatible with panoptic segmentation and additionally they have part-level labels for selected semantic classes. This report describes the format of the two datasets, the annotation protocols, the merging strategies, and presents the datasets statistics. The datasets labels together with code for processing and visualization will be published at https://github.com/tue-mps/panoptic_parts.

cs.CV↗

arXiv2020

Semantic Foreground Inpainting from Weak Supervision

Semantic scene understanding is an essential task for self-driving vehicles and mobile robots. In our work, we aim to estimate a semantic segmentation map, in which the foreground objects are removed and semantically inpainted with background classes, from a single RGB image. This semantic foreground inpainting task is performed by a single-stage convolutional neural network (CNN) that contains our novel max-pooling as inpainting (MPI) module, which is trained with weak supervision, i.e., it does not require manual background annotations for the foreground regions to be inpainted. Our approach is inherently more efficient than the previous two-stage state-of-the-art method, and outperforms it by a margin of 3% IoU for the inpainted foreground regions on Cityscapes. The performance margin increases to 6% IoU, when tested on the unseen KITTI dataset. The code and the manually annotated datasets for testing are shared with the research community at https://github.com/Chenyang-Lu/semantic-foreground-inpainting.

cs.CV↗

arXiv2019

Hallucinating Beyond Observation: Learning to Complete with Partial Observation and Unpaired Prior Knowledge

We propose a novel single-step training strategy that allows convolutional encoder-decoder networks that use skip connections, to complete partially observed data by means of hallucination. This strategy is demonstrated for the task of completing 2-D road layouts as well as 3-D vehicle shapes. As input, it takes data from a partially observed domain, for which no ground truth is available, and data from an unpaired prior knowledge domain and trains the network in an end-to-end manner. Our single-step training strategy is compared against two state-of-the-art baselines, one using a two-step auto-encoder training strategy and one using an adversarial strategy. Our novel strategy achieves an improvement up to +12.2% F-measure on the Cityscapes dataset. The learned network intrinsically generalizes better than the baselines on unseen datasets, which is demonstrated by an improvement up to +23.8% F-measure on the unseen KITTI dataset. Moreover, our approach outperforms the baselines using the same backbone network on the 3-D shape completion benchmark by a margin of 0.006 Hamming distance.

cs.CV↗

arXiv2018

Monocular Semantic Occupancy Grid Mapping with Convolutional Variational Encoder-Decoder Networks

In this work, we research and evaluate end-to-end learning of monocular semantic-metric occupancy grid mapping from weak binocular ground truth. The network learns to predict four classes, as well as a camera to bird's eye view mapping. At the core, it utilizes a variational encoder-decoder network that encodes the front-view visual information of the driving scene and subsequently decodes it into a 2-D top-view Cartesian coordinate system. The evaluations on Cityscapes show that the end-to-end learning of semantic-metric occupancy grids outperforms the deterministic mapping approach with flat-plane assumption by more than 12% mean IoU. Furthermore, we show that the variational sampling with a relatively small embedding vector brings robustness against vehicle dynamic perturbations, and generalizability for unseen KITTI data. Our network achieves real-time inference rates of approx. 35 Hz for an input image with a resolution of 256x512 pixels and an output map with 64x64 occupancy grid cells using a Titan V GPU.

cs.RO↗

arXiv2018

Predicting Clinical Deterioration of Outpatients Using Multimodal Data Collected by Wearables

Hospital readmission rate is high for heart failure patients. Early detection of deterioration will help doctors prevent readmissions, thus reducing health care cost and providing patients with just-in-time intervention. Wearable devices (e.g., wristbands and smart watches) provide a convenient technology for continuous outpatient monitoring. In the paper, we explore the feasibility of monitoring outpatients using Fitbit Charge HR wristbands and the potential of machine learning models to predicting clinical deterioration (readmissions and death) among outpatients discharged from the hospital. We developed and piloted a data collection system in a clinical study which involved 25 heart failure patients recently discharged from a hospital. The results from the clinical study demonstrated the feasibility of continuously monitoring outpatients using wristbands. We observed high levels of patient compliance in wearing the wristbands regularly and satisfactory yield, latency and reliability of data collection from the wristbands to a cloud-based database. Finally, we explored a set of machine learning models to predict deterioration based on the Fitbit data. Through 5-fold cross validation, K nearest neighbor achieved the highest accuracy of 0.8800 for identifying patients at risk of deterioration using the health data from the beginning of the monitoring. Machine learning models based on multimodal data (step, sleep and heart rate) significantly outperformed the traditional clinical approach based on LACE index. Moreover, our proposed weighted samples one class SVM model can reach high accuracy (0.9635) for predicting the deterioration happening in the future using data collected by a sliding window, which indicates the potential for allowing timely intervention.

cs.CY↗