arXiv subjects

Zhen Xu

Publications and source records attributed to Zhen Xu.

At least 109 records · Page 6Linked to original sources

arXiv2022

Comparison of Spatio-Temporal Models for Human Motion and Pose Forecasting in Face-to-Face Interaction Scenarios

Human behavior forecasting during human-human interactions is of utmost importance to provide robotic or virtual agents with social intelligence. This problem is especially challenging for scenarios that are highly driven by interpersonal dynamics. In this work, we present the first systematic comparison of state-of-the-art approaches for behavior forecasting. To do so, we leverage whole-body annotations (face, body, and hands) from the very recently released UDIVA v0.5, which features face-to-face dyadic interactions. Our best attention-based approaches achieve state-of-the-art performance in UDIVA v0.5. We show that by autoregressively predicting the future with methods trained for the short-term future (<400ms), we outperform the baselines even for a considerably longer-term future (up to 2s). We also show that this finding holds when highly noisy annotations are used, which opens new horizons towards the use of weakly-supervised learning. Combined with large-scale datasets, this may help boost the advances in this field.

cs.CV↗

arXiv2022

Didn't see that coming: a survey on non-verbal social human behavior forecasting

Non-verbal social human behavior forecasting has increasingly attracted the interest of the research community in recent years. Its direct applications to human-robot interaction and socially-aware human motion generation make it a very attractive field. In this survey, we define the behavior forecasting problem for multiple interactive agents in a generic way that aims at unifying the fields of social signals prediction and human motion forecasting, traditionally separated. We hold that both problem formulations refer to the same conceptual problem, and identify many shared fundamental challenges: future stochasticity, context awareness, history exploitation, etc. We also propose a taxonomy that comprises methods published in the last 5 years in a very informative way and describes the current main concerns of the community with regard to this problem. In order to promote further research on this field, we also provide a summarised and friendly overview of audiovisual datasets featuring non-acted social interactions. Finally, we describe the most common metrics used in this task and their particular issues.

cs.CV↗

arXiv2022

Codabench: Flexible, Easy-to-Use and Reproducible Benchmarking Platform

Obtaining standardized crowdsourced benchmark of computational methods is a major issue in data science communities. Dedicated frameworks enabling fair benchmarking in a unified environment are yet to be developed. Here we introduce Codabench, an open-source, community-driven platform for benchmarking algorithms or software agents versus datasets or tasks. A public instance of Codabench (https://www.codabench.org/) is open to everyone, free of charge, and allows benchmark organizers to compare fairly submissions, under the same setting (software, hardware, data, algorithms), with custom protocols and data formats. Codabench has unique features facilitating the organization of benchmarks flexibly, easily and reproducibly, such as the possibility of re-using templates of benchmarks, and supplying compute resources on-demand. Codabench has been used internally and externally on various applications, receiving more than 130 users and 2500 submissions. As illustrative use cases, we introduce 4 diverse benchmarks covering Graph Machine Learning, Cancer Heterogeneity, Clinical Diagnosis and Reinforcement Learning.

cs.LG↗

arXiv2022

Winning solutions and post-challenge analyses of the ChaLearn AutoDL challenge 2019

This paper reports the results and post-challenge analyses of ChaLearn's AutoDL challenge series, which helped sorting out a profusion of AutoML solutions for Deep Learning (DL) that had been introduced in a variety of settings, but lacked fair comparisons. All input data modalities (time series, images, videos, text, tabular) were formatted as tensors and all tasks were multi-label classification problems. Code submissions were executed on hidden tasks, with limited time and computational resources, pushing solutions that get results quickly. In this setting, DL methods dominated, though popular Neural Architecture Search (NAS) was impractical. Solutions relied on fine-tuned pre-trained networks, with architectures matching data modality. Post-challenge tests did not reveal improvements beyond the imposed time limit. While no component is particularly original or novel, a high level modular organization emerged featuring a "meta-learner", "data ingestor", "model selector", "model/learner", and "evaluator". This modularity enabled ablation studies, which revealed the importance of (off-platform) meta-learning, ensembling, and efficient data management. Experiments on heterogeneous module combinations further confirm the (local) optimality of the winning solutions. Our challenge legacy includes an ever-lasting benchmark (http://autodl.chalearn.org), the open-sourced code of the winners, and a free "AutoDL self-service".

cs.LG↗

arXiv2021

AutoML Meets Time Series Regression Design and Analysis of the AutoSeries Challenge

Analyzing better time series with limited human effort is of interest to academia and industry. Driven by business scenarios, we organized the first Automated Time Series Regression challenge (AutoSeries) for the WSDM Cup 2020. We present its design, analysis, and post-hoc experiments. The code submission requirement precluded participants from any manual intervention, testing automated machine learning capabilities of solutions, across many datasets, under hardware and time limitations. We prepared 10 datasets from diverse application domains (sales, power consumption, air quality, traffic, and parking), featuring missing data, mixed continuous and categorical variables, and various sampling rates. Each dataset was split into a training and a test sequence (which was streamed, allowing models to continuously adapt). The setting of time series regression, differs from classical forecasting in that covariates at the present time are known. Great strides were made by participants to tackle this AutoSeries problem, as demonstrated by the jump in performance from the sample submission, and post-hoc comparisons with AutoGluon. Simple yet effective methods were used, based on feature engineering, LightGBM, and random search hyper-parameter tuning, addressing all aspects of the challenge. Our post-hoc analyses revealed that providing additional time did not yield significant improvements. The winners' code was open-sourced https://github.com/NehzUx/AutoSeries.

cs.LG↗

arXiv2021

MUFASA: Multimodal Fusion Architecture Search for Electronic Health Records

One important challenge of applying deep learning to electronic health records (EHR) is the complexity of their multimodal structure. EHR usually contains a mixture of structured (codes) and unstructured (free-text) data with sparse and irregular longitudinal features -- all of which doctors utilize when making decisions. In the deep learning regime, determining how different modality representations should be fused together is a difficult problem, which is often addressed by handcrafted modeling and intuition. In this work, we extend state-of-the-art neural architecture search (NAS) methods and propose MUltimodal Fusion Architecture SeArch (MUFASA) to simultaneously search across multimodal fusion strategies and modality-specific architectures for the first time. We demonstrate empirically that our MUFASA method outperforms established unimodal NAS on public EHR data with comparable computation costs. In addition, MUFASA produces architectures that outperform Transformer and Evolved Transformer. Compared with these baselines on CCS diagnosis code prediction, our discovered models improve top-5 recall from 0.88 to 0.91 and demonstrate the ability to generalize to other EHR tasks. Studying our top architecture in depth, we provide empirical evidence that MUFASA's improvements are derived from its ability to both customize modeling for each data modality and find effective fusion strategies.

cs.LG↗

arXiv2021

Layer-dependent Optical and Dielectric Properties of Large-size PdSe$_2$ Films Grown by Chemical Vapor Deposition

Palladium diselenide (PdSe$_2$), a new type of two-dimensional noble metal dihalides (NMDCs), has received widespread attention for its excellent electrical and optoelectronic properties. Herein, high-quality continuous centimeter-scale PdSe$_2$ films with layers in the range of 3L-15L were grown using Chemical Vapor Deposition (CVD) method. The absorption spectra and DFT calculations revealed that the bandgap of the PdSe$_2$ films decreased with increasing number of layers, which is due to PdSe$_2$ enhancement of orbital hybridization. Spectroscopic ellipsometry (SE) analysis shows that PdSe2 has significant layer-dependent optical and dielectric properties. This is mainly due to the unique strong exciton effect of the thin PdSe$_2$ film in the UV band. In particular, the effect of temperature on the optical properties of PdSe$_2$ films was also observed, and the thermo-optical coefficients of PdSe$_2$ films with different number of layers were calculated. This study provides fundamental guidance for the fabrication and optimization of PdSe$_2$-based optoelectronic devices.

cond-mat.mtrl-sci↗

arXiv2021

Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020

This paper presents the results and insights from the black-box optimization (BBO) challenge at NeurIPS 2020 which ran from July-October, 2020. The challenge emphasized the importance of evaluating derivative-free optimizers for tuning the hyperparameters of machine learning models. This was the first black-box optimization challenge with a machine learning emphasis. It was based on tuning (validation set) performance of standard machine learning models on real datasets. This competition has widespread impact as black-box optimization (e.g., Bayesian optimization) is relevant for hyperparameter tuning in almost every machine learning project as well as many applications outside of machine learning. The final leaderboard was determined using the optimization performance on held-out (hidden) objective functions, where the optimizers ran without human intervention. Baselines were set using the default settings of several open-source black-box optimization packages as well as random search.

cs.LG↗

arXiv2021

Physics-based Constitutive Modeling of Photo-oxidative Aging in Semi-Crystalline Polymers based on Chemical Characterization Techniques

This paper proposes a physio-chemically-based constitutive framework to simulate and predict the response of semi-crystalline low-density polyethylene (LDPE) to severe photo-oxidation. Photo-oxidation induced by exposure to Ultra-Violet (UV) light and oxygen is the dominant degradation mechanism affecting the lifespan of LDPE. In this work, we propose evolution functions for the material properties in the constitutive equations of \cite{boyce2000constitutive} to incorporate the effects of photo-oxidation on the mechanical response of LDPE. The evolution functions are based on chemically verified processes that are responsible for material degradation, namely the change in crystallinity and mass loss relative to the initial pristine films over exposure time. Changes in crystallinity and mass loss are characterized by Differential Scanning Calorimetry (DSC) and Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D) experiments, respectively. Connecting the physio-chemical processes affecting polymer network evolution to the mechanical response of LDPE bypasses the need for defining fitting parameters that carry no physical meaning. The developed constitutive framework is validated with respect to a series of in-house uniaxial tensile tests performed on LDPE aged for different UV exposure times. Comparison of the constitutive framework versus experimental mechanical tests also confirms the accuracy of DSC and QCM-D as rigorous techniques to monitor and characterize degradation in LDPE films. The outcome shed light on the evolution of the macromolecular network in LDPE under extreme photo-oxidation and the evolution of the associated mechanical material properties.

cond-mat.mtrl-sci↗

arXiv2021

BEDS-Bench: Behavior of EHR-models under Distributional Shift--A Benchmark

Machine learning has recently demonstrated impressive progress in predictive accuracy across a wide array of tasks. Most ML approaches focus on generalization performance on unseen data that are similar to the training data (In-Distribution, or IND). However, real world applications and deployments of ML rarely enjoy the comfort of encountering examples that are always IND. In such situations, most ML models commonly display erratic behavior on Out-of-Distribution (OOD) examples, such as assigning high confidence to wrong predictions, or vice-versa. Implications of such unusual model behavior are further exacerbated in the healthcare setting, where patient health can potentially be put at risk. It is crucial to study the behavior and robustness properties of models under distributional shift, understand common failure modes, and take mitigation steps before the model is deployed. Having a benchmark that shines light upon these aspects of a model is a first and necessary step in addressing the issue. Recent work and interest in increasing model robustness in OOD settings have focused more on image modality, while the Electronic Health Record (EHR) modality is still largely under-explored. We aim to bridge this gap by releasing BEDS-Bench, a benchmark for quantifying the behavior of ML models over EHR data under OOD settings. We use two open access, de-identified EHR datasets to construct several OOD data settings to run tests on, and measure relevant metrics that characterize crucial aspects of a model's OOD behavior. We evaluate several learning algorithms under BEDS-Bench and find that all of them show poor generalization performance under distributional shift in general. Our results highlight the need and the potential to improve robustness of EHR models under distributional shift, and BEDS-Bench provides one way to measure progress towards that goal.

cs.LG↗

arXiv2021

MatchVIE: Exploiting Match Relevancy between Entities for Visual Information Extraction

Visual Information Extraction (VIE) task aims to extract key information from multifarious document images (e.g., invoices and purchase receipts). Most previous methods treat the VIE task simply as a sequence labeling problem or classification problem, which requires models to carefully identify each kind of semantics by introducing multimodal features, such as font, color, layout. But simply introducing multimodal features couldn't work well when faced with numeric semantic categories or some ambiguous texts. To address this issue, in this paper we propose a novel key-value matching model based on a graph neural network for VIE (MatchVIE). Through key-value matching based on relevancy evaluation, the proposed MatchVIE can bypass the recognitions to various semantics, and simply focuses on the strong relevancy between entities. Besides, we introduce a simple but effective operation, Num2Vec, to tackle the instability of encoded values, which helps model converge more smoothly. Comprehensive experiments demonstrate that the proposed MatchVIE can significantly outperform previous methods. Notably, to the best of our knowledge, MatchVIE may be the first attempt to tackle the VIE task by modeling the relevancy between keys and values and it is a good complement to the existing methods.

cs.CV↗

arXiv2021

NeoRL: A Near Real-World Benchmark for Offline Reinforcement Learning

Offline reinforcement learning (RL) aims at learning a good policy from a batch of collected data, without extra interactions with the environment during training. However, current offline RL benchmarks commonly have a large reality gap, because they involve large datasets collected by highly exploratory policies, and the trained policy is directly evaluated in the environment. In real-world situations, running a highly exploratory policy is prohibited to ensure system safety, the data is commonly very limited, and a trained policy should be well validated before deployment. In this paper, we present a near real-world offline RL benchmark, named NeoRL, which contains datasets from various domains with controlled sizes, and extra test datasets for policy validation. We evaluate existing offline RL algorithms on NeoRL and argue that the performance of a policy should also be compared with the deterministic version of the behavior policy, instead of the dataset reward. The empirical results demonstrate that the tested offline RL algorithms become less competitive to the deterministic policy on many datasets, and the offline policy evaluation hardly helps. The NeoRL suit can be found at http://polixir.ai/research/neorl. We hope this work will shed some light on future research and draw more attention when deploying RL in real-world systems.

cs.LG↗

arXiv2021

Growth and Collapse of an Isolated Bubble Driven by a Single Negative Histotripsy Cycle in Agarose Gel: Stress, Strain, and Strain Rate Fields

Histotripsy relies on cavitation to mechanically homogenize soft tissue. There is strong evidence that the high stresses, strains, and strain rates developed as bubbles grow and collapse contribute to this tissue homogenization. While such stresses and strains have been examined computationally in model systems with assumed constitutive models (e.g., finite-deformation Neo-Hookean model) and viscoelastic properties determined under quasi-static conditions, recent studies proposed that the Quadratic Law Kelvin-Voigt (QLKV) constitutive model, which additionally accounts for strain stiffening, more accurately represents the viscoelastic response of soft materials subjected to cavitation; this model has also been used to infer viscoelastic properties at high rates. In this work, we use the QLKV model and these properties to calculate the time-dependent stress, strain, and strain rate fields produced during the growth and collapse of individual bubbles subjected to a histotripsy-relevant pressure waveform in agarose gels of 0.3~\% and 1.0~\% concentration and corresponding to actual (past) experiments. We find that, as the gel concentration is increased, strain stiffening manifests in larger elastic stresses and compressive stresses extending into the collapse phase, particularly for the 1.0~\% concentration gel. As a result, the duration of the collapse phase also increases. In comparison with the conventional Neo-Hookean model, the compressive stress has a larger magnitude, extends farther into the surrounding medium, and shows an increased departure from growth/collapse symmetry close to the bubble; all of these effects are magnified in the stiffer gel.

cond-mat.soft↗

arXiv2021

Solution landscapes of the diblock copolymer-homopolymer model under two-dimensional confinement

We investigate the solution landscapes of the confined diblock copolymer and homopolymer in two-dimensional domain by using the extended Ohta--Kawasaki model. The projected saddle dynamics method is developed to compute the saddle points with mass conservation and construct the solution landscape by coupling with downward/upward search algorithms. A variety of novel stationary solutions are identified and classified in the solution landscape, including Flower class, Mosaic class, Core-shell class, and Tai-chi class. The relationships between different stable states are shown by either transition pathways connected by index-1 saddle points or dynamical pathways connected by a high-index saddle point. The solution landscapes also demonstrate the symmetry-breaking phenomena, in which more solutions with high symmetry are found when the domain size increases.

cond-mat.soft↗

arXiv2020

Acoustic Cavitation Rheometry

Characterization of soft materials is challenging due to their high compliance and the strain-rate dependence of their mechanical properties. The inertial microcavitation-based high strain-rate rheometry (IMR) method [Estrada et al., J. Mech. Phys. Solids, 2018, 112, 291-317] combines laser-induced cavitation measurements with a model for the bubble dynamics to measure local properties of polyacrylamide hydrogel under high strain-rates from $10^3$ to $10^8$ s$^{-1}$. While promising, laser-induced cavitation involves plasma formation and optical breakdown during nucleation, a process that could alter local material properties before measurements are obtained. In the present study, we extend the IMR method to another means to generate cavitation, namely high-amplitude focused ultrasound, and apply the resulting acoustic-cavitation-based IMR to characterize the mechanical properties of agarose hydrogels. Material properties including viscosity, elastic constants, and a stress-free bubble radius are inferred from bubble radius histories in $0.3\%$ and $1\%$ agarose gels. An ensemble-based data assimilation is used to further help interpret the obtained estimates. The resulting parameter distributions are consistent with available measurements of agarose gel properties and with expected trends related to gel concentration and high strain-rate loading. Our findings demonstrate the utility of applying IMR and data assimilation methods with single-bubble acoustic cavitation data for measurement of viscoelastic properties.

cond-mat.soft↗

arXiv2020

AutoSpeech 2020: The Second Automated Machine Learning Challenge for Speech Classification

The AutoSpeech challenge calls for automated machine learning (AutoML) solutions to automate the process of applying machine learning to speech processing tasks. These tasks, which cover a large variety of domains, will be shown to the automated system in a random order. Each time when the tasks are switched, the information of the new task will be hinted with its corresponding training set. Thus, every submitted solution should contain an adaptation routine which adapts the system to the new task. Compared to the first edition, the 2020 edition includes advances of 1) more speech tasks, 2) noisier data in each task, 3) a modified evaluation metric. This paper outlines the challenge and describe the competition protocol, datasets, evaluation metric, starting kit, and baseline systems.

cs.AI↗

arXiv2020

Online Learning with Cumulative Oversampling: Application to Budgeted Influence Maximization

We propose a cumulative oversampling (CO) method for online learning. Our key idea is to sample parameter estimations from the updated belief space once in each round (similar to Thompson Sampling), and utilize the cumulative samples up to the current round to construct optimistic parameter estimations that asymptotically concentrate around the true parameters as tighter upper confidence bounds compared to the ones constructed with standard UCB methods. We apply CO to a novel budgeted variant of the Influence Maximization (IM) semi-bandits with linear generalization of edge weights, whose offline problem is NP-hard. Combining CO with the oracle we design for the offline problem, our online learning algorithm simultaneously tackles budget allocation, parameter learning, and reward maximization. We show that for IM semi-bandits, our CO-based algorithm achieves a scaled regret comparable to that of the UCB-based algorithms in theory, and performs on par with Thompson Sampling in numerical experiments.

cs.LG↗

arXiv2020

Single-bubble dynamics in histotripsy and high-amplitude ultrasound: Modeling and validation

A variety of approaches have been used to model the dynamics of a single, isolated bubble nucleated by a microsecond length high-amplitude ultrasound pulse (e.g., a histotripsy pulse). Until recently, the lack of single--bubble experimental radius vs. time data for bubble dynamics under a well-characterized driving pressure has limited model validation efforts. This study uses radius vs. time measurements of single, spherical histotripsy-nucleated bubbles in water [Wilson et al., Phys. Rev. E, 2019, 99, 043103] to quantitatively compare and validate a variety of bubble dynamics modeling approaches, including compressible and incompressible models as well as different thermal models. A strategy for inferring an analytic representation of histotripsy waveforms directly from experimental radius vs. time and cavitation threshold data is presented. We compare distributions of a calculated validation metric obtained for each model applied to $88$ experimental data sets. There is minimal distinction ($< 1\%$) among the modeling approaches for compressibility and thermal effects considered in this study. These results suggest that our proposed strategy to infer the waveform, combined with simple models minimizing parametric uncertainty and computational resource demands accurately represent single-bubble dynamics in histotripsy, including at and near the maximum bubble radius. Remaining sources of parametric and model-based uncertainty are discussed.

physics.flu-dyn↗