Understanding the parameter estimation of softmax gating Gaussian mixture of experts has remained a long-standing open problem in the literature. It is mainly due to three fundamental theoretical challenges associated with the softmax gating function: (i) the identifiability only up to the translation of parameters; (ii) the intrinsic interaction via partial differential equations between the softmax gating and the expert functions in the Gaussian density; (iii) the complex dependence between the numerator and denominator of the conditional density of softmax gating Gaussian mixture of experts. We resolve these challenges by proposing novel Voronoi loss functions among parameters and establishing the convergence rates of maximum likelihood estimator (MLE) for solving parameter estimation in these models. When the true number of experts is unknown and over-specified, our findings show a connection between the convergence rate of the MLE and a solvability problem of a system of polynomial equations.
相關內容
It has been shown that learning audiovisual features can lead to improved speech recognition performance over audio-only features, especially for noisy speech. However, in many common applications, the visual features are partially or entirely missing, e.g.~the speaker might move off screen. Multi-modal models need to be robust: missing video frames should not degrade the performance of an audiovisual model to be worse than that of a single-modality audio-only model. While there have been many attempts at building robust models, there is little consensus on how robustness should be evaluated. To address this, we introduce a framework that allows claims about robustness to be evaluated in a precise and testable way. We also conduct a systematic empirical study of the robustness of common audiovisual speech recognition architectures on a range of acoustic noise conditions and test suites. Finally, we show that an architecture-agnostic solution based on cascades can consistently achieve robustness to missing video, even in settings where existing techniques for robustness like dropout fall short.
Multiphysics incompressible fluid dynamics simulations play a crucial role in understanding intricate behaviors of many complex engineering systems that involve interactions between solids, fluids, and various phases like liquid and gas. Numerical modeling of these interactions has generated significant research interest in recent decades and has led to the development of open source simulation tools and commercial software products targeting specific applications or general problem classes in computational fluid dynamics. As the demand increases for these simulations to adapt to platform heterogeneity, ensure composability between different physics models, and effectively utilize inheritance within partial differentiation systems, a fundamental reconsideration of numerical solver design becomes imperative. The discussion presented in this paper emphasizes the importance of these considerations and introduces the Flash-X approach as a potential solution. The software design strategies outlined in the article serve as a guide for Flash-X developers, providing insights into complexities associated with performance portability, composability, and sustainable development. These strategies provide a foundation for improving design of both new and existing simulation tools grappling with these challenges. By incorporating the principles outlined in the Flash-X approach, engineers and researchers can enhance the adaptability, efficiency, and overall effectiveness of their numerical solvers in the ever-evolving field of multiphysics simulations.
Real-world decision-making problems are usually accompanied by delayed rewards, which affects the sample efficiency of Reinforcement Learning, especially in the extremely delayed case where the only feedback is the episodic reward obtained at the end of an episode. Episodic return decomposition is a promising way to deal with the episodic-reward setting. Several corresponding algorithms have shown remarkable effectiveness of the learned step-wise proxy rewards from return decomposition. However, these existing methods lack either attribution or representation capacity, leading to inefficient decomposition in the case of long-term episodes. In this paper, we propose a novel episodic return decomposition method called Diaster (Difference of implicitly assigned sub-trajectory reward). Diaster decomposes any episodic reward into credits of two divided sub-trajectories at any cut point, and the step-wise proxy rewards come from differences in expectation. We theoretically and empirically verify that the decomposed proxy reward function can guide the policy to be nearly optimal. Experimental results show that our method outperforms previous state-of-the-art methods in terms of both sample efficiency and performance.
Kinship verification from face images is a novel and formidable challenge in the realms of pattern recognition and computer vision. This work makes notable contributions by incorporating a preprocessing technique known as Multiscale Retinex (MSR), which enhances image quality. Our approach harnesses the strength of complementary deep (VGG16) and shallow texture descriptors (BSIF) by combining them at the score level using Logistic Regression (LR) technique. We assess the effectiveness of our approach by conducting comprehensive experiments on three challenging kinship datasets: Cornell Kin Face, UB Kin Face and TS Kin Face
With the rapid progress in virtual reality (VR) technology, the scope of VR applications has greatly expanded across various domains. However, the superiority of VR training over traditional methods and its impact on learning efficacy are still uncertain. To investigate whether VR training is more effective than traditional methods, we designed virtual training systems for mechanical assembly on both VR and desktop platforms, subsequently conducting pre-test and post-test experiments. A cohort of 53 students, all enrolled in engineering drawing course without prior knowledge distinctions, was randomly divided into three groups: physical training, desktop virtual training, and immersive VR training. Our investigation utilized analysis of covariance (ANCOVA) to examine the differences in post-test scores among the three groups while controlling for pre-test scores. The group that received VR training showed the highest scores on the post-test. Another facet of our study delved into the presence of the virtual system. We developed a specialized scale to assess this aspect for our research objectives. Our findings indicate that VR training can enhance the sense of presence, particularly in terms of sensory factors and realism factors. Moreover, correlation analysis uncovers connections between the various dimensions of presence. This study confirms that using VR training can improve learning efficacy and the presence in the context of mechanical assembly, surpassing traditional training methods. Furthermore, it provides empirical evidence supporting the integration of VR technology in higher education and engineering training. This serves as a reference for the practical application of VR technology in different fields.
Particle Swarm Optimization (PSO) has emerged as a powerful metaheuristic global optimization approach over the past three decades. Its appeal lies in its ability to tackle complex multidimensional problems that defy conventional algorithms. However, PSO faces challenges, such as premature stagnation in single-objective scenarios and the need to strike a balance between exploration and exploitation. Hybridizing PSO by integrating its cooperative nature with established optimization techniques from diverse paradigms offers a promising solution. In this paper, we investigate various strategies for synergizing gradient-based optimizers with PSO. We introduce different hybridization principles and explore several approaches, including sequential decoupled hybridization, coupled hybridization, and adaptive hybridization. These strategies aim to enhance the efficiency and effectiveness of PSO, ultimately improving its ability to navigate intricate optimization landscapes. By combining the strengths of gradient-based methods with the inherent social dynamics of PSO, we seek to address the critical objectives of intelligent exploration and exploitation in complex optimization tasks. Our study delves into the comparative merits of these hybridization techniques and offers insights into their application across different problem domains.
In the global context, while mixed reality has been an emerging concept for years, recent technological and scientific advancements have now made it poised to revolutionize industries and daily life by offering enhanced functionalities and improved services. Besides reviewing the highly cited papers in the last 20 years among over a thousand research papers on mixed reality, this systematic review provides the state-of-the-art applications and utilities of the mixed reality by primarily scrutinizing the associated papers in 2022 and 2023. Focusing on the potentials that this technology have in providing digitally supported simulations and other utilities in the era of large language models, highlighting the potential and limitations of the innovative solutions and also bringing focus to emerging research directions, such as telemedicine, remote control and optimization of direct volume rendering. The paper's associated repository is publicly accessible at //aizierjiang.github.io/mr.
Graph Neural Networks (GNNs) demonstrate their significance by effectively modeling complex interrelationships within graph-structured data. To enhance the credibility and robustness of GNNs, it becomes exceptionally crucial to bolster their ability to capture causal relationships. However, despite recent advancements that have indeed strengthened GNNs from a causal learning perspective, conducting an in-depth analysis specifically targeting the causal modeling prowess of GNNs remains an unresolved issue. In order to comprehensively analyze various GNN models from a causal learning perspective, we constructed an artificially synthesized dataset with known and controllable causal relationships between data and labels. The rationality of the generated data is further ensured through theoretical foundations. Drawing insights from analyses conducted using our dataset, we introduce a lightweight and highly adaptable GNN module designed to strengthen GNNs' causal learning capabilities across a diverse range of tasks. Through a series of experiments conducted on both synthetic datasets and other real-world datasets, we empirically validate the effectiveness of the proposed module.
We investigate two efficient time discretizations for the post-processing technique of discontinuous Galerkin (DG) methods to solve hyperbolic conservation laws. The post-processing technique, which is applied at the final time of the DG method, can enhance the accuracy of the original DG solution (spatial superconvergence). One main difficulty of the post-processing technique is that the spatial superconvergence after post-processing needs to be matched with proper temporary accuracy. If the semi-discretized system (ODE system after spatial discretization) is under-resolved in time, then the space superconvergence will be concealed. In this paper, we focus our investigation on the recently designed SDG method and derive its explicit scheme from a correction process based on the DG weak formulation. We also introduce another similar technique, namely the spectral deferred correction (SDC) method. A comparison is made among both proposed time discretization techniques with the standard third-order Runge-Kutta method through several numerical examples, to conclude that both the SDG and SDC methods are efficient time discretization techniques for exploiting the spatial superconvergence of the DG methods.
We present the design of a mixed reality (MR) telehealth training system that aims to close the gap between in-person and distance training and re-training for medical procedures. Our system uses real-time volumetric capture as a means for communicating and relating spatial information between the non-colocated trainee and instructor. The system's design is based on a requirements elicitation study performed in situ, at a medical school simulation training center. The focus is on the lightweight real-time transmission of volumetric data - meaning the use of consumer hardware, easy and quick deployment, and low-demand computations. We evaluate the MR system design by analyzing the workload for the users during medical training. We compare in-person, video, and MR training workloads. The results indicate that the overall workload for central line placement training with MR does not increase significantly compared to video communication. Our work shows that, when designed strategically together with domain experts, an MR communication system can be used effectively for complex medical procedural training without increasing the overall workload for users significantly. Moreover, MR systems offer new opportunities for teaching due to spatial information, hand tracking, and augmented communication.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191