arXiv subjects

Lingxiao Wang

Publications and source records attributed to Lingxiao Wang.

At least 91 records · Page 5Linked to original sources

arXiv2021

Revisiting Membership Inference Under Realistic Assumptions

We study membership inference in settings where some of the assumptions typically used in previous research are relaxed. First, we consider skewed priors, to cover cases such as when only a small fraction of the candidate pool targeted by the adversary are actually members and develop a PPV-based metric suitable for this setting. This setting is more realistic than the balanced prior setting typically considered by researchers. Second, we consider adversaries that select inference thresholds according to their attack goals and develop a threshold selection procedure that improves inference attacks. Since previous inference attacks fail in imbalanced prior setting, we develop a new inference attack based on the intuition that inputs corresponding to training set members will be near a local minimum in the loss function, and show that an attack that combines this with thresholds on the per-instance loss can achieve high PPV even in settings where other attacks appear to be ineffective. Code for our experiments can be found here: https://github.com/bargavj/EvaluatingDPML.

cs.CR↗

arXiv2020

Efficient and Robust Propensity-Score-Based Methods for Population Inference using Epidemiologic Cohorts

Most epidemiologic cohorts are composed of volunteers who do not represent the general population. To enable population inference from cohorts, we and others have proposed utilizing probability survey samples as external references to develop a propensity score (PS) for membership in the cohort versus survey. Herein we develop a unified framework for PS-based weighting (such as inverse PS weighting (IPSW)) and matching methods (such as kernel-weighting (KW) method). We identify a fundamental Strong Exchangeability Assumption (SEA) underlying existing PS-based matching methods whose failure invalidates inference even if the PS-model is correctly specified. We relax the SEA to a Weak Exchangeability Assumption (WEA) for the matching method. Also, we propose IPSW.S and KW.S methods that reduce the variance of PS-based estimators by scaling the survey weights used in the PS estimation. We prove consistency of the IPSW.S and KW.S estimators of population means and prevalences under WEA, and provide asymptotic variances and consistent variance estimators. In simulations, the KW.S and IPSW.S estimators had smallest MSE. In our data example, the original KW estimates had large bias, whereas the KW.S estimates had the smallest MSE.

stat.ME↗

arXiv2020

Machine learning spatio-temporal epidemiological model to evaluate Germany-county-level COVID-19 risk

As the COVID-19 pandemic continues to ravage the world, it is of critical significance to provide a timely risk prediction of the COVID-19 in multi-level. To implement it and evaluate the public health policies, we develop a framework with machine learning assisted to extract epidemic dynamics from the infection data, in which contains a county-level spatiotemporal epidemiological model that combines a spatial Cellular Automaton (CA) with a temporal Susceptible-Undiagnosed-Infected-Removed (SUIR) model. Compared with the existing time risk prediction models, the proposed CA-SUIR model shows the multi-level risk of the county to the government and coronavirus transmission patterns under different policies. This new toolbox is first utilized to the projection of the multi-level COVID-19 prevalence over 412 Landkreis (counties) in Germany, including t-day-ahead risk forecast and the risk assessment to the travel restriction policy. As a practical illustration, we predict the situation at Christmas where the worst fatalities are 34.5 thousand, effective policies could contain it to below 21 thousand. Such intervenable evaluation system could help decide on economic restarting and public health policies making in pandemic.

physics.soc-ph↗

arXiv2020

Continuous-mixture Autoregressive Networks for efficient variational calculation of many-body systems

We develop deep autoregressive networks with multi channels to compute many-body systems with \emph{continuous} spin degrees of freedom directly. As a concrete example, we embed the two-dimensional XY model into the continuous-mixture networks and rediscover the Kosterlitz-Thouless (KT) phase transition on a periodic square lattice. Vortices characterizing the quasi-long range order are accurately detected by the autoregressive neural networks. By learning the microscopic probability distributions from the macroscopic thermal distribution, the neural networks compute the free energy directly and find that free vortices and anti-vortices emerge in the high-temperature regime. As a more precise evaluation, we compute the helicity modulus to determine the KT transition temperature. Although the training process becomes more time-consuming with larger lattice sizes, the training time remains unchanged around the KT transition temperature. The continuous-mixture autoregressive networks we developed thus can be potentially used to study other many-body systems with continuous degrees of freedom.

cond-mat.dis-nn↗

arXiv2020

Neural Network Statistical Mechanics

We propose a general framework to extract microscopic interactions from raw configurations with deep neural networks. The approach replaces the modeling Hamiltonian by the neural networks, in which the interaction is encoded. It can be trained with data collected from Ab initio computations or experiments. The well-trained neural networks give an accurate estimation of the possibility distribution of the configurations at fixed external parameters. It can be spontaneously extrapolated to detect the phase structures since classical statistical mechanics as prior knowledge here. We apply the approach to a 2D spin system, training at a fixed temperature, and reproducing the phase structure. Scaling the configuration on lattice exhibits the interaction changes with the degree of freedom, which can be naturally applied to the experimental measurements. Our approach bridges the gap between the real configurations and the microscopic dynamics with an autoregressive neural network.

physics.comp-ph↗

arXiv2020

Mode Decomposed Chiral Magnetic Effect and Rotating Fermions

We present a novel perspective to characterize the chiral magnetic and related effects in terms of angular decomposed modes. We find that the vector current and the chirality density are connected through a surprisingly simple relation for all the modes and any mass, which defines the mode decomposed chiral magnetic effect in such a way free from the chiral chemical potential. The mode decomposed formulation is useful also to investigate properties of rotating fermions. For demonstration we give an intuitive account for a nonzero density emerging from a combination of rotation and magnetic field as well as an approach to the chiral vortical effect at finite density.

hep-ph↗

arXiv2020

On the Global Optimality of Model-Agnostic Meta-Learning

Model-agnostic meta-learning (MAML) formulates meta-learning as a bilevel optimization problem, where the inner level solves each subtask based on a shared prior, while the outer level searches for the optimal shared prior by optimizing its aggregated performance over all the subtasks. Despite its empirical success, MAML remains less understood in theory, especially in terms of its global optimality, due to the nonconvexity of the meta-objective (the outer-level objective). To bridge such a gap between theory and practice, we characterize the optimality gap of the stationary points attained by MAML for both reinforcement learning and supervised learning, where the inner-level and outer-level problems are solved via first-order optimization methods. In particular, our characterization connects the optimality gap of such stationary points with (i) the functional geometry of inner-level objectives and (ii) the representation power of function approximators, including linear models and neural networks. To the best of our knowledge, our analysis establishes the global optimality of MAML with nonconvex meta-objectives for the first time.

cs.LG↗

arXiv2020

Provably Efficient Causal Reinforcement Learning with Confounded Observational Data

Empowered by expressive function approximators such as neural networks, deep reinforcement learning (DRL) achieves tremendous empirical successes. However, learning expressive function approximators requires collecting a large dataset (interventional data) by interacting with the environment. Such a lack of sample efficiency prohibits the application of DRL to critical scenarios, e.g., autonomous driving and personalized medicine, since trial and error in the online setting is often unsafe and even unethical. In this paper, we study how to incorporate the dataset (observational data) collected offline, which is often abundantly available in practice, to improve the sample efficiency in the online setting. To incorporate the possibly confounded observational data, we propose the deconfounded optimistic value iteration (DOVI) algorithm, which incorporates the confounded observational data in a provably efficient manner. More specifically, DOVI explicitly adjusts for the confounding bias in the observational data, where the confounders are partially observed or unobserved. In both cases, such adjustments allow us to construct the bonus based on a notion of information gain, which takes into account the amount of information acquired from the offline setting. In particular, we prove that the regret of DOVI is smaller than the optimal regret achievable in the pure online setting by a multiplicative factor, which decreases towards zero when the confounded observational data are more informative upon the adjustments. Our algorithm and analysis serve as a step towards causal reinforcement learning.

cs.LG↗

arXiv2020

Breaking the Curse of Many Agents: Provable Mean Embedding Q-Iteration for Mean-Field Reinforcement Learning

Multi-agent reinforcement learning (MARL) achieves significant empirical successes. However, MARL suffers from the curse of many agents. In this paper, we exploit the symmetry of agents in MARL. In the most generic form, we study a mean-field MARL problem. Such a mean-field MARL is defined on mean-field states, which are distributions that are supported on continuous space. Based on the mean embedding of the distributions, we propose MF-FQI algorithm that solves the mean-field MARL and establishes a non-asymptotic analysis for MF-FQI algorithm. We highlight that MF-FQI algorithm enjoys a "blessing of many agents" property in the sense that a larger number of observed agents improves the performance of MF-FQI algorithm.

cs.LG↗

arXiv2019

Neural Policy Gradient Methods: Global Optimality and Rates of Convergence

Policy gradient methods with actor-critic schemes demonstrate tremendous empirical successes, especially when the actors and critics are parameterized by neural networks. However, it remains less clear whether such "neural" policy gradient methods converge to globally optimal policies and whether they even converge at all. We answer both the questions affirmatively in the overparameterized regime. In detail, we prove that neural natural policy gradient converges to a globally optimal policy at a sublinear rate. Also, we show that neural vanilla policy gradient converges sublinearly to a stationary point. Meanwhile, by relating the suboptimality of the stationary points to the representation power of neural actor and critic classes, we prove the global optimality of all stationary points under mild regularity conditions. Particularly, we show that a key to the global optimality and convergence is the "compatibility" between the actor and critic, which is ensured by sharing neural architectures and random initializations across the actor and critic. To the best of our knowledge, our analysis establishes the first global optimality and convergence guarantees for neural policy gradient methods.

cs.LG↗

arXiv2019

Chiral vortices and pseudoscalar condensation due to rotation

We investigate the influence of rotation on the dynamical chiral symmetry breaking in strongly interacting matter. We develop a self-consistent Bogoliubov-de Gennes-like theoretical framework to study the inhomogeneous chiral condensate and the possible chiral vortex state in rotating finite-size matter in four-fermion interacting theories. We show that for sufficiently rapid rotation in $2+1$ dimensions, the ground state can be a chiral vortex state, a type of topological defect in analogy to superfluids and superconductors. The vortex state exhibits pion condensation, providing a new mechanism to realize pseudoscalar condensation in strongly interacting matter.

nucl-th↗

arXiv2019

A Knowledge Transfer Framework for Differentially Private Sparse Learning

We study the problem of estimating high dimensional models with underlying sparse structures while preserving the privacy of each training example. We develop a differentially private high-dimensional sparse learning framework using the idea of knowledge transfer. More specifically, we propose to distill the knowledge from a "teacher" estimator trained on a private dataset, by creating a new dataset from auxiliary features, and then train a differentially private "student" estimator using this new dataset. In addition, we establish the linear convergence rate as well as the utility guarantee for our proposed method. For sparse linear regression and sparse logistic regression, our method achieves improved utility guarantees compared with the best known results (Kifer et al., 2012; Wang and Gu, 2019). We further demonstrate the superiority of our framework through both synthetic and real-world data experiments.

stat.ML↗

arXiv2019

Escape dynamics based on bounded rationality

The bounded rationality plays a vital role in the collective behavior of the evacuation process. Also investigating human behavior in such an extreme situation is a continuing concern within social psychology. In this paper, we construct a cellular automaton (CA) model for the escape dynamics, and the bounded rational behavior induced by heterogeneous information is introduced. The non-trivial behavior shows in the replicator dynamics method with mean field approximation, where people's perception of the distribution of population and velocity is reduced to an average value in a certain direction. Analyzing the escape efficiency shows that under the premise of rationality, the bounded rational strategy can get higher performance. Interestingly, a quantifiable meta-stable state appears in the escape process, and the escape time is power-law dependent on system size.

physics.soc-ph↗

arXiv2018

Local suppression and enhancement of pairing condensate under rotation

The rotation induced inhomogeneous problem is non-trivial and inevitable. In this paper a generic framework is developed to investigate the inhomogeneous condensate in a system of fermions under the presence of rotation. It is a self-consistent method basing on a set of relativistic BdG equations solved with typical iteration algorithm. Taking the chiral condensate for example we study rotational effects numerically and discover two inhomogeneous effects, the local rotational suppression effect and centrifugal effect. They may have significant impacts on the phase structure of various kinds of matter. Several systems in different physics branches have been discussed in the paper.

nucl-th↗

arXiv2018

Learning One-hidden-layer ReLU Networks via Gradient Descent

We study the problem of learning one-hidden-layer neural networks with Rectified Linear Unit (ReLU) activation function, where the inputs are sampled from standard Gaussian distribution and the outputs are generated from a noisy teacher network. We analyze the performance of gradient descent for training such kind of neural networks based on empirical risk minimization, and provide algorithm-dependent guarantees. In particular, we prove that tensor initialization followed by gradient descent can converge to the ground-truth parameters at a linear rate up to some statistical error. To the best of our knowledge, this is the first work characterizing the recovery guarantee for practical learning of one-hidden-layer ReLU networks with multiple neurons. Numerical experiments verify our theoretical findings.

stat.ML↗

arXiv2018

Nambu--Jona-Lasinio model in a parallel electromagnetic field

We explore the features of the $U_A(1)$ and chiral symmetry breaking of the Nambu--Jona-Lasinio model without the Kobayashi-Maskawa-'t Hooft determinant term in the presence of a parallel electromagnetic field. We show that the electromagnetic chiral anomaly can induce both finite neutral pion condensate and isospin-singlet pseudo-scalar $η$ condensate and thus modifies the chiral symmetry breaking pattern. In order to characterize the strength of the $U_A(1)$ symmetry breaking, we evaluate the susceptibility associated with the $U_A(1)$ charge. The result shows that the susceptibility contributed from the chiral anomaly is consistent with the behavior of the corresponding $η$ condensate. The spectra of the mesonic excitations are also studied.

nucl-th↗

arXiv2018

A Unified Framework for Low-Rank plus Sparse Matrix Recovery

We propose a unified framework to solve general low-rank plus sparse matrix recovery problems based on matrix factorization, which covers a broad family of objective functions satisfying the restricted strong convexity and smoothness conditions. Based on projected gradient descent and the double thresholding operator, our proposed generic algorithm is guaranteed to converge to the unknown low-rank and sparse matrices at a locally linear rate, while matching the best-known robustness guarantee (i.e., tolerance for sparsity). At the core of our theory is a novel structural Lipschitz gradient condition for low-rank plus sparse matrices, which is essential for proving the linear convergence rate of our algorithm, and we believe is of independent interest to prove fast rates for general superposition-structured models. We illustrate the application of our framework through two concrete examples: robust matrix sensing and robust PCA. Experiments on both synthetic and real datasets corroborate our theory.

stat.ML↗

arXiv2018

Competition between magnetic catalysis effect and chiral rotation effect

In this work, we explore the competition between magnetic catalysis effect and chiral rotation effect in a general parallel electromagnetic field within the effective Nambu--Jona-Lasinio model. For a given electric field $E$ at zero temperature, the mass gap shows three different features with respect to an increasing magnetic field $B$: increasing monotonically, decreasing after increasing and decreasing monotonically. By making use of strong magnetic field approximation, we illuminate that this is due to the competition between catalysis effect and chiral rotation effect induced both by the magnetic field, and a critical electric field $\sqrt{eE_c}=86.4~{\rm MeV}$ is found beyond which the mass gap will eventually decrease at large $B$. As only large magnetic field is relevant for the derivation, the critical electric field does not depend on the temperature $T$ or chemical potential $μ$.

nucl-th↗