arXiv subjects

Jiacheng Zhu

Publications and source records attributed to Jiacheng Zhu.

70 records · Page 4Linked to original sources

arXiv2022

PhysioMTL: Personalizing Physiological Patterns using Optimal Transport Multi-Task Regression

Heart rate variability (HRV) is a practical and noninvasive measure of autonomic nervous system activity, which plays an essential role in cardiovascular health. However, using HRV to assess physiology status is challenging. Even in clinical settings, HRV is sensitive to acute stressors such as physical activity, mental stress, hydration, alcohol, and sleep. Wearable devices provide convenient HRV measurements, but the irregularity of measurements and uncaptured stressors can bias conventional analytical methods. To better interpret HRV measurements for downstream healthcare applications, we learn a personalized diurnal rhythm as an accurate physiological indicator for each individual. We develop Physiological Multitask-Learning (PhysioMTL) by harnessing Optimal Transport theory within a Multitask-learning (MTL) framework. The proposed method learns an individual-specific predictive model from heterogeneous observations, and enables estimation of an optimal transport map that yields a push forward operation onto the demographic features for each task. Our model outperforms competing MTL methodologies on unobserved predictive tasks for synthetic and two real-world datasets. Specifically, our method provides remarkable prediction results on unseen held-out subjects given only $20\%$ of the subjects in real-world observational studies. Furthermore, our model enables a counterfactual engine that generates the effect of acute stressors and chronic conditions on HRV rhythms.

eess.SP↗

arXiv2021

Quantum oscillations in 2D insulators induced by graphite gates

We demonstrate a mechanism for magnetoresistance oscillations in insulating states of two-dimensional (2D) materials arising from the interaction of the 2D layer and proximal graphite gates. We study a series of devices based on different two-dimensional systems, including mono- and bilayer Td-WTe2, angle-aligned MoTe2/WSe2 heterobilayers and Bernal-stacked bilayer graphene, which all share a similar graphite-gated geometry. We find that the resistivity of the 2D system generically shows quantum oscillations as a function of magnetic field corresponding to a high-density Fermi surface when they are tuned near an insulating state, in contravention of naïve band theory. Simultaneous measurement of the resistivity of the graphite gates show that these oscillations are precisely correlated with quantum oscillations in the resistivity of the graphite gates themselves. Further supporting this connection, the oscillations are quenched when the graphite gate is replaced by TaSe2, a high-density metal that does not show quantum oscillations. The observed phenomenon arises from the oscillatory behavior of graphite density of states, which modulates the device capacitance and, as a consequence, the carrier density in the sample layer even when a constant electrochemical potential is maintained between the sample and the gate electrode. Oscillations are most pronounced near insulating states where the resistivity is strongly density dependent. Our study suggests a unified mechanism for quantum oscillations in graphite-gated 2D insulators based on sample-gate coupling.

cond-mat.mes-hall↗

arXiv2021

Continuous Mott transition in semiconductor moiré superlattices

The evolution of a Landau Fermi liquid into a nonmagnetic Mott insulator with increasing electronic interactions is one of the most puzzling quantum phase transitions in physics. The vicinity of the transition is believed to host exotic states of matter such as quantum spin liquids, exciton condensates and unconventional superconductivity. Semiconductor moiré materials realize a highly controllable Hubbard model simulator on a triangular lattice, providing a unique opportunity to drive a metal-insulator transition (MIT) via continuous tuning of the electronic interactions. Here, by electrically tuning the effective interaction strength in MoTe2/WSe2 moiré superlattices, we observe a continuous MIT at a fixed filling of one electron per unit cell. The existence of quantum criticality is supported by the scaling behavior of the resistance, a continuously vanishing charge-gap as the critical point is approached from the insulating side, and a diverging quasiparticle effective mass from the metallic side. We also observe a smooth evolution of the low-temperature magnetic susceptibility across the MIT and find no evidence of long-range magnetic order down to ~ 5% of the Curie-Weiss temperature. The results signal an abundance of low-energy spinful excitations on the insulating side that is further corroborated by the presence of the Pomeranchuk effect on the metallic side. Our results are consistent with the universal critical theory of a continuous MIT from a Landau Fermi liquid to a nonmagnetic Mott insulator in two dimensions.

cond-mat.str-el↗

arXiv2021

Charge-order-enhanced capacitance in semiconductor moiré superlattices

Van der Waals moiré materials have emerged as a highly controllable platform to study the electronic correlation phenomena. In particular, robust correlated insulating states have recently been discovered at both integer and fractional filling factors of the semiconductor moiré systems. Here, we reveal the thermodynamic properties of these states by measuring the gate capacitance of MoSe2/WS2 moiré superlattices. We observe a series of incompressible states for filling factor 0 - 8 and anomalously large capacitance (nearly 60% above the device's geometrical capacitance) in the intervening compressible regions. The anomalously large capacitance is most pronounced at small filling factor, below the melting temperature of the charge-ordered states, and for small sample-gate separation. It is a manifestation of the device-geometry-dependent Coulomb interaction between electrons and phase mixing of the charge-ordered states. We have further extracted the thermodynamic gap of the correlated insulating states and the entropy of the capacitive device. The results not only establish capacitance as a powerful probe of the correlated states in semiconductor moiré systems, but also demonstrate control of the extended Coulomb interaction in these materials via sample-gate coupling.

cond-mat.str-el↗

arXiv2021

Context-Aware Safe Reinforcement Learning for Non-Stationary Environments

Safety is a critical concern when deploying reinforcement learning agents for realistic tasks. Recently, safe reinforcement learning algorithms have been developed to optimize the agent's performance while avoiding violations of safety constraints. However, few studies have addressed the non-stationary disturbances in the environments, which may cause catastrophic outcomes. In this paper, we propose the context-aware safe reinforcement learning (CASRL) method, a meta-learning framework to realize safe adaptation in non-stationary environments. We use a probabilistic latent variable model to achieve fast inference of the posterior environment transition distribution given the context data. Safety constraints are then evaluated with uncertainty-aware trajectory sampling. The high cost of safety violations leads to the rareness of unsafe records in the dataset. We address this issue by enabling prioritized sampling during model training and formulating prior safety constraints with domain knowledge during constrained planning. The algorithm is evaluated in realistic safety-critical environments with non-stationary disturbances. Results show that the proposed algorithm significantly outperforms existing baselines in terms of safety and robustness.

cs.LG↗

arXiv2020

Task-Agnostic Online Reinforcement Learning with an Infinite Mixture of Gaussian Processes

Continuously learning to solve unseen tasks with limited experience has been extensively pursued in meta-learning and continual learning, but with restricted assumptions such as accessible task distributions, independently and identically distributed tasks, and clear task delineations. However, real-world physical tasks frequently violate these assumptions, resulting in performance degradation. This paper proposes a continual online model-based reinforcement learning approach that does not require pre-training to solve task-agnostic problems with unknown task boundaries. We maintain a mixture of experts to handle nonstationarity, and represent each different type of dynamics with a Gaussian Process to efficiently leverage collected data and expressively model uncertainty. We propose a transition prior to account for the temporal dependencies in streaming data and update the mixture online via sequential variational inference. Our approach reliably handles the task distribution shift by generating new models for never-before-seen dynamics and reusing old models for previously seen dynamics. In experiments, our approach outperforms alternative methods in non-stationary tasks, including classic control with changing dynamics and decision making in different driving scenarios.

cs.LG↗

arXiv2020

Spatiotemporal Learning of Multivehicle Interaction Patterns in Lane-Change Scenarios

Interpretation of common-yet-challenging interaction scenarios can benefit well-founded decisions for autonomous vehicles. Previous research achieved this using their prior knowledge of specific scenarios with predefined models, limiting their adaptive capabilities. This paper describes a Bayesian nonparametric approach that leverages continuous (i.e., Gaussian processes) and discrete (i.e., Dirichlet processes) stochastic processes to reveal underlying interaction patterns of the ego vehicle with other nearby vehicles. Our model relaxes dependency on the number of surrounding vehicles by developing an acceleration-sensitive velocity field based on Gaussian processes. The experiment results demonstrate that the velocity field can represent the spatial interactions between the ego vehicle and its surroundings. Then, a discrete Bayesian nonparametric model, integrating Dirichlet processes and hidden Markov models, is developed to learn the interaction patterns over the temporal space by segmenting and clustering the sequential interaction data into interpretable granular patterns automatically. We then evaluate our approach in the highway lane-change scenarios using the highD dataset collected from real-world settings. Results demonstrate that our proposed Bayesian nonparametric approach provides an insight into the complicated lane-change interactions of the ego vehicle with multiple surrounding traffic participants based on the interpretable interaction patterns and their transition properties in temporal relationships. Our proposed approach sheds light on efficiently analyzing other kinds of multi-agent interactions, such as vehicle-pedestrian interactions. View the demos via https://youtu.be/z_vf9UHtdAM.

cs.RO↗

arXiv2020

Stripe phases in WSe2/WS2 moiré superlattices

Stripe phases, in which the rotational symmetry of charge density is spontaneously broken, occur in many strongly correlated systems with competing interactions. One representative example is the copper-oxide superconductors, where stripe order is thought to be relevant to the mechanism of high-temperature superconductivity. Identifying and studying the stripe phases in conventional strongly correlated systems are, however, challenging due to the complexity and limited tunability of these materials. Here we uncover stripe phases in WSe2/WS2 moiré superlattices with continuously gate-tunable charge densities by combining optical anisotropy and electronic compressibility measurements. We find strong electronic anisotropy over a large doping range peaked at 1/2 filling of the moiré superlattice. The 1/2-state is incompressible and assigned to a (insulating) stripe crystal phase. It can be continuously melted by thermal fluctuations around 35 K. The domain configuration revealed by wide-field imaging shows a preferential alignment along the high-symmetry axes of the moiré superlattice. Away from 1/2 filling, we observe additional stripe crystals at commensurate filling 1/4, 2/5 and 3/5. The anisotropy also extends into the compressible regime of the system at incommensurate fillings, indicating the presence of electronic liquid crystal states. The observed filling-dependent stripe phases agree with the theoretical phase diagram of the extended Hubbard model on a triangular lattice in the flat band limit. Our results demonstrate that two-dimensional semiconductor moiré superlattices are a highly tunable platform to study the stripe phases and their interplay with other symmetry breaking ground states.

cond-mat.mes-hall↗

arXiv2020

Creation of moiré bands in a monolayer semiconductor by spatially periodic dielectric screening

Moiré superlattices of two-dimensional van der Waals materials have emerged as a powerful platform for designing electronic band structures and discovering emergent physical phenomena. A key concept involves the creation of long-wavelength periodic potential and moiré bands in a crystal through interlayer hybridization when two materials are overlaid. Here we demonstrate a new approach based on spatially periodic dielectric screening to create moiré bands in a monolayer semiconductor. It relies on reduced dielectric screening of the Coulomb interactions in monolayer semiconductors and their environmental dielectric-dependent electronic band structure. We observe optical transitions between moiré bands in monolayer WSe$_{2}$ when it is placed close to small angle-misaligned graphene on hexagonal boron nitride. The moiré bands are a result of long-range Coulomb interactions, strongly gate-tunable, and can have versatile superlattice symmetries independent of the crystal lattice of the host material. Our result also demonstrates that monolayer semiconductors are sensitive local dielectric sensors.

cond-mat.mtrl-sci↗

arXiv2020

Robust Unsupervised Learning of Temporal Dynamic Interactions

Robust representation learning of temporal dynamic interactions is an important problem in robotic learning in general and automated unsupervised learning in particular. Temporal dynamic interactions can be described by (multiple) geometric trajectories in a suitable space over which unsupervised learning techniques may be applied to extract useful features from raw and high-dimensional data measurements. Taking a geometric approach to robust representation learning for temporal dynamic interactions, it is necessary to develop suitable metrics and a systematic methodology for comparison and for assessing the stability of an unsupervised learning method with respect to its tuning parameters. Such metrics must account for the (geometric) constraints in the physical world as well as the uncertainty associated with the learned patterns. In this paper we introduce a model-free metric based on the Procrustes distance for robust representation learning of interactions, and an optimal transport based distance metric for comparing between distributions of interaction primitives. These distance metrics can serve as an objective for assessing the stability of an interaction learning algorithm. They are also used for comparing the outcomes produced by different algorithms. Moreover, they may also be adopted as an objective function to obtain clusters and representative interaction primitives. These concepts and techniques will be introduced, along with mathematical properties, while their usefulness will be demonstrated in unsupervised learning of vehicle-to-vechicle interactions extracted from the Safety Pilot database, the world's largest database for connected vehicles.

cs.LG↗

arXiv2019

Modeling Multi-Vehicle Interaction Scenarios Using Gaussian Random Field

Autonomous vehicles are expected to navigate in complex traffic scenarios with multiple surrounding vehicles. The correlations between road users vary over time, the degree of which, in theory, could be infinitely large, thus posing a great challenge in modeling and predicting the driving environment. In this paper, we propose a method to model multi-vehicle interactions using a stochastic vector field model and apply non-parametric Bayesian learning to extract the underlying motion patterns from a large quantity of naturalistic traffic data. We then use this model to reproduce the high-dimensional driving scenarios in a finitely tractable form. We use a Gaussian process to model multi-vehicle motion, and a Dirichlet process to assign each observation to a specific scenario. We verify the effectiveness of the proposed method on highway and intersection datasets from the NGSIM project, in which complex multi-vehicle interactions are prevalent. The results show that the proposed method can capture motion patterns from both settings, without imposing heroic prior, and hence demonstrate the potential application for a wide array of traffic situations. The proposed modeling method could enable simulation platforms and other testing methods designed for autonomous vehicle evaluation, to easily model and generate traffic scenarios emulating large scale driving data.

cs.RO↗

arXiv2019

Probabilistic Trajectory Prediction for Autonomous Vehicles with Attentive Recurrent Neural Process

Predicting surrounding vehicle behaviors are critical to autonomous vehicles when negotiating in multi-vehicle interaction scenarios. Most existing approaches require tedious training process with large amounts of data and may fail to capture the propagating uncertainty in interaction behaviors. The multi-vehicle behaviors are assumed to be generated from a stochastic process. This paper proposes an attentive recurrent neural process (ARNP) approach to overcome the above limitations, which uses a neural process (NP) to learn a distribution of multi-vehicle interaction behavior. Our proposed model inherits the flexibility of neural networks while maintaining Bayesian probabilistic characteristics. Constructed by incorporating NPs with recurrent neural networks (RNNs), the ARNP model predicts the distribution of a target vehicle trajectory conditioned on the observed long-term sequential data of all surrounding vehicles. This approach is verified by learning and predicting lane-changing trajectories in complex traffic scenarios. Experimental results demonstrate that our proposed method outperforms previous counterparts in terms of accuracy and uncertainty expressiveness. Moreover, the meta-learning instinct of NPs enables our proposed ARNP model to capture global information of all observations, thereby being able to adapt to new targets efficiently.

cs.RO↗

arXiv2019

Recurrent Attentive Neural Process for Sequential Data

Neural processes (NPs) learn stochastic processes and predict the distribution of target output adaptively conditioned on a context set of observed input-output pairs. Furthermore, Attentive Neural Process (ANP) improved the prediction accuracy of NPs by incorporating attention mechanism among contexts and targets. In a number of real-world applications such as robotics, finance, speech, and biology, it is critical to learn the temporal order and recurrent structure from sequential data. However, the capability of NPs capturing these properties is limited due to its permutation invariance instinct. In this paper, we proposed the Recurrent Attentive Neural Process (RANP), or alternatively, Attentive Neural Process-RecurrentNeural Network(ANP-RNN), in which the ANP is incorporated into a recurrent neural network. The proposed model encapsulates both the inductive biases of recurrent neural networks and also the strength of NPs for modelling uncertainty. We demonstrate that RANP can effectively model sequential data and outperforms NPs and LSTMs remarkably in a 1D regression toy example as well as autonomous-driving applications.

cs.LG↗

arXiv2019

A General Framework of Learning Multi-Vehicle Interaction Patterns from Videos

Semantic learning and understanding of multi-vehicle interaction patterns in a cluttered driving environment are essential but challenging for autonomous vehicles to make proper decisions. This paper presents a general framework to gain insights into intricate multi-vehicle interaction patterns from bird's-eye view traffic videos. We adopt a Gaussian velocity field to describe the time-varying multi-vehicle interaction behaviors and then use deep autoencoders to learn associated latent representations for each temporal frame. Then, we utilize a hidden semi-Markov model with a hierarchical Dirichlet process as a prior to segment these sequential representations into granular components, also called traffic primitives, corresponding to interaction patterns. Experimental results demonstrate that our proposed framework can extract traffic primitives from videos, thus providing a semantic way to analyze multi-vehicle interaction patterns, even for cluttered driving scenarios that are far messier than human beings can cope with.

cs.RO↗

arXiv2019

Estimating Risk Levels of Driving Scenarios through Analysis of Driving Styles for Autonomous Vehicles

In order to operate safely on the road, autonomous vehicles need not only to be able to identify objects in front of them, but also to be able to estimate the risk level of the object in front of the vehicle automatically. It is obvious that different objects have different levels of danger to autonomous vehicles. An evaluation system is needed to automatically determine the danger level of the object for the autonomous vehicle. It would be too subjective and incomplete if the system were completely defined by humans. Based on this, we propose a framework based on nonparametric Bayesian learning method -- a sticky hierarchical Dirichlet process hidden Markov model(sticky HDP-HMM), and discover the relationship between driving scenarios and driving styles. We use the analysis of driving styles of autonomous vehicles to reflect the risk levels of driving scenarios to the vehicles. In this framework, we firstly use sticky HDP-HMM to extract driving styles from the dataset and get different clusters, then an evaluation system is proposed to evaluate and rank the urgency levels of the clusters. Finally, we map the driving scenarios to the ranking results and thus get clusters of driving scenarios in different risk levels. More importantly, we find the relationship between driving scenarios and driving styles. The experiment shows that our framework can cluster and rank driving styles of different urgency levels and find the relationship between driving scenarios and driving styles and the conclusions also fit people's common sense when driving. Furthermore, this framework can be used for autonomous vehicles to estimate risk levels of driving scenarios and help them make precise and safe decisions.

cs.RO↗

arXiv2018

A Tempt to Unify Heterogeneous Driving Databases using Traffic Primitives

A multitude of publicly-available driving datasets and data platforms have been raised for autonomous vehicles (AV). However, the heterogeneities of databases in size, structure and driving context make existing datasets practically ineffective due to a lack of uniform frameworks and searchable indexes. In order to overcome these limitations on existing public datasets, this paper proposes a data unification framework based on traffic primitives with ability to automatically unify and label heterogeneous traffic data. This is achieved by two steps: 1) Carefully arrange raw multidimensional time series driving data into a relational database and then 2) automatically extract labeled and indexed traffic primitives from traffic data through a Bayesian nonparametric learning method. Finally, we evaluate the effectiveness of our developed framework using the collected real vehicle data.

cs.CV↗