The transportation sector significantly contributes to greenhouse gas emissions, necessitating accurate emission models guide mitigation strategies. Despite its field validation and certification, the industry-standard Motor Vehicle Emission Simulator (MOVES) faces challenges related complexity in usage, high computational demands, unsuitability for microscopic real-time applications. To address these limitations, we present NeuralMOVES, a comprehensive suite of high-performance, lightweight...

This paper examines a continuous-time routing system with general interarrival and service time distributions, operating under either the join-the-shortest-queue policy or power-of-two-choices policy. Under weaker set of assumptions than those commonly established in literature, we prove that scaled steady-state queue length at each station converges weakly to common exponential random variable heavy traffic. Specifically, our results hold assumption (2 + ε)th moment for distributions some ε...

A prominent concern of scientific investigators is the presence unobserved hidden variables in association analysis. Ignoring often yields biased statistical results and misleading conclusions. Motivated by this practical issue, paper studies multivariate response regression with variables, Y=(Ψ∗)TX+(B∗)TZ+E, where Y∈Rm vector, X∈Rp observable feature, Z∈RK represents vector possibly correlated X, E an independent error. The number K unknown both m p are allowed, but not required, to grow...

In temporal-difference reinforcement learning algorithms, variance in value estimation can cause instability and overestimation of the maximal target value. Many algorithms have been proposed to reduce overestimation, including several recent ensemble methods, however none shown success sample-efficient through addressing as root overestimation. this paper, we propose MeanQ, a simple method that estimates values means. Despite its simplicity, MeanQ shows remarkable sample efficiency...

This paper studies the continuous-time join-the-shortest-queue (JSQ) system with general interarrival and service distributions. Under a much weaker assumption than one in literature, we prove that each station's scaled steady-state queue length weakly converges to an identical exponential random variable heavy traffic. Specifically, establish our results by only assuming $2+\delta_0$ moment on arrival distributions for some $\delta_0>0$. Our proof exploits Palm version of basic adjoint...

The weighted-workload-task-allocation (WWTA) load-balancing policy is known to be throughput optimal for parallel server systems with heterogeneous servers. This work concerns the heavy traffic approximation of steady-state performance operating under WWTA policy. Under a relaxed complete-resource-pooling condition, we prove that achieves "strong form" state-space collapse in and scaled workload each converges distribution an exponential random variable, whose parameter explicitly given by...

This paper studies the continuous-time join-the-shortestqueue (JSQ) system with general interarrival and service distributions.

Soft Actor-Critic (SAC) is considered the state-of-the-art algorithm in continuous action space settings. It uses maximum entropy framework for efficiency and stability, applies a heuristic temperature Lagrange term to tune $\alpha$, which determines how "soft" policy should be. counter-intuitive that empirical evidence shows SAC does not perform well discrete domains. In this paper we investigate possible explanations phenomenon propose Target Entropy Scheduled (TES-SAC), an annealing...

Temporal-Difference (TD) learning methods, such as Q-Learning, have proven effective at a policy to perform control tasks. One issue with methods like Q-Learning is that the value update introduces bias when predicting TD target of unfamiliar state. Estimation noise becomes after max operator in improvement step, and carries over estimations other states, causing overestimate Q value. Algorithms Soft (SQL) introduce notion soft-greedy policy, which reduces estimation via soft updates early...

We prove that under a multi-scale heavy traffic condition, the stationary distribution of scaled queue length vector process in any generalized Jackson network has product-form limit. Each component product form an exponential distribution, corresponding to Brownian approximation single station queue. The ``single station'' can be constructed precisely and its parameters have good intuitive interpretation.

This paper develops an algorithm to predict the number of Covid-19 patients who will start use ventilators tomorrow. is intended be utilized by a large hospital or group coordinated hospitals at end each day (e.g. 8pm) when current non-ventilated and predicated admissions for tomorrow are available. The predicted new can replaced numbers in previous d days (including today) some integer ≥ 1 such data In our simulation model that calibrated with New York City's data, predictions have...

This paper studies the estimation of coefficient matrix $\Ttheta$ in multivariate regression with hidden variables, $Y = (\Ttheta)^TX + (B^*)^TZ E$, where $Y$ is a $m$-dimensional response vector, $X$ $p$-dimensional vector observable features, $Z$ represents $K$-dimensional unobserved possibly correlated $X$, and $E$ an independent error. The number variables $K$ unknown both $m$ $p$ are allowed but not required to grow sample size $n$. Since only observable, we provide necessary conditions...