arXiv subjects

Ziyi Chen

Publications and source records attributed to Ziyi Chen.

85 records · Page 5Linked to original sources

arXiv2022

A Fast and Convergent Proximal Algorithm for Regularized Nonconvex and Nonsmooth Bi-level Optimization

Many important machine learning applications involve regularized nonconvex bi-level optimization. However, the existing gradient-based bi-level optimization algorithms cannot handle nonconvex or nonsmooth regularizers, and they suffer from a high computation complexity in nonconvex bi-level optimization. In this work, we study a proximal gradient-type algorithm that adopts the approximate implicit differentiation (AID) scheme for nonconvex bi-level optimization with possibly nonconvex and nonsmooth regularizers. In particular, the algorithm applies the Nesterov's momentum to accelerate the computation of the implicit gradient involved in AID. We provide a comprehensive analysis of the global convergence properties of this algorithm through identifying its intrinsic potential function. In particular, we formally establish the convergence of the model parameters to a critical point of the bi-level problem, and obtain an improved computation complexity $\mathcal{O}(κ^{3.5}ε^{-2})$ over the state-of-the-art result. Moreover, we analyze the asymptotic convergence rates of this algorithm under a class of local nonconvex geometries characterized by a Łojasiewicz-type gradient inequality. Experiment on hyper-parameter optimization demonstrates the effectiveness of our algorithm.

cs.LG↗

arXiv2022

Accelerated Proximal Alternating Gradient-Descent-Ascent for Nonconvex Minimax Machine Learning

Alternating gradient-descent-ascent (AltGDA) is an optimization algorithm that has been widely used for model training in various machine learning applications, which aims to solve a nonconvex minimax optimization problem. However, the existing studies show that it suffers from a high computation complexity in nonconvex minimax optimization. In this paper, we develop a single-loop and fast AltGDA-type algorithm that leverages proximal gradient updates and momentum acceleration to solve regularized nonconvex minimax optimization problems. By leveraging the momentum acceleration technique, we prove that the algorithm converges to a critical point in nonconvex minimax optimization and achieves a computation complexity in the order of $\mathcal{O}(κ^{\frac{11}{6}}ε^{-2})$, where $ε$ is the desired level of accuracy and $κ$ is the problem's condition number. {Such a computation complexity improves the state-of-the-art complexities of single-loop GDA and AltGDA algorithms (see the summary of comparison in \Cref{table1})}. We demonstrate the effectiveness of our algorithm via an experiment on adversarial deep learning.

cs.LG↗

arXiv2022

Transformer-Based Multi-Aspect Multi-Granularity Non-Native English Speaker Pronunciation Assessment

Automatic pronunciation assessment is an important technology to help self-directed language learners. While pronunciation quality has multiple aspects including accuracy, fluency, completeness, and prosody, previous efforts typically only model one aspect (e.g., accuracy) at one granularity (e.g., at the phoneme-level). In this work, we explore modeling multi-aspect pronunciation assessment at multiple granularities. Specifically, we train a Goodness Of Pronunciation feature-based Transformer (GOPT) with multi-task learning. Experiments show that GOPT achieves the best results on speechocean762 with a public automatic speech recognition (ASR) acoustic model trained on Librispeech.

cs.SD↗

arXiv2022

Data Sampling Affects the Complexity of Online SGD over Dependent Data

Conventional machine learning applications typically assume that data samples are independently and identically distributed (i.i.d.). However, practical scenarios often involve a data-generating process that produces highly dependent data samples, which are known to heavily bias the stochastic optimization process and slow down the convergence of learning. In this paper, we conduct a fundamental study on how different stochastic data sampling schemes affect the sample complexity of online stochastic gradient descent (SGD) over highly dependent data. Specifically, with a $ϕ$-mixing model of data dependence, we show that online SGD with proper periodic data-subsampling achieves an improved sample complexity over the standard online SGD in the full spectrum of the data dependence level. Interestingly, even subsampling a subset of data samples can accelerate the convergence of online SGD over highly dependent data. Moreover, we show that online SGD with mini-batch sampling can further substantially improve the sample complexity over online SGD with periodic data-subsampling over highly dependent data. Numerical experiments validate our theoretical results.

cs.LG↗

arXiv2022

Sample and Communication-Efficient Decentralized Actor-Critic Algorithms with Finite-Time Analysis

Actor-critic (AC) algorithms have been widely adopted in decentralized multi-agent systems to learn the optimal joint control policy. However, existing decentralized AC algorithms either do not preserve the privacy of agents or are not sample and communication-efficient. In this work, we develop two decentralized AC and natural AC (NAC) algorithms that are private, and sample and communication-efficient. In both algorithms, agents share noisy information to preserve privacy and adopt mini-batch updates to improve sample and communication efficiency. Particularly for decentralized NAC, we develop a decentralized Markovian SGD algorithm with an adaptive mini-batch size to efficiently compute the natural policy gradient. Under Markovian sampling and linear function approximation, we prove the proposed decentralized AC and NAC algorithms achieve the state-of-the-art sample complexities $\mathcal{O}\big(ε^{-2}\ln(ε^{-1})\big)$ and $\mathcal{O}\big(ε^{-3}\ln(ε^{-1})\big)$, respectively, and the same small communication complexity $\mathcal{O}\big(ε^{-1}\ln(ε^{-1})\big)$. Numerical experiments demonstrate that the proposed algorithms achieve lower sample and communication complexities than the existing decentralized AC algorithm.

cs.LG↗

arXiv2022

The HCCL-DKU system for fake audio generation task of the 2022 ICASSP ADD Challenge

The voice conversion task is to modify the speaker identity of continuous speech while preserving the linguistic content. Generally, the naturalness and similarity are two main metrics for evaluating the conversion quality, which has been improved significantly in recent years. This paper presents the HCCL-DKU entry for the fake audio generation task of the 2022 ICASSP ADD challenge. We propose a novel ppg-based voice conversion model that adopts a fully end-to-end structure. Experimental results show that the proposed method outperforms other conversion models, including Tacotron-based and Fastspeech-based models, on conversion quality and spoofing performance against anti-spoofing systems. In addition, we investigate several post-processing methods for better spoofing power. Finally, we achieve second place with a deception success rate of 0.916 in the ADD challenge.

cs.SD↗

arXiv2021

Greedy-GQ with Variance Reduction: Finite-time Analysis and Improved Complexity

Greedy-GQ is a value-based reinforcement learning (RL) algorithm for optimal control. Recently, the finite-time analysis of Greedy-GQ has been developed under linear function approximation and Markovian sampling, and the algorithm is shown to achieve an $ε$-stationary point with a sample complexity in the order of $\mathcal{O}(ε^{-3})$. Such a high sample complexity is due to the large variance induced by the Markovian samples. In this paper, we propose a variance-reduced Greedy-GQ (VR-Greedy-GQ) algorithm for off-policy optimal control. In particular, the algorithm applies the SVRG-based variance reduction scheme to reduce the stochastic variance of the two time-scale updates. We study the finite-time convergence of VR-Greedy-GQ under linear function approximation and Markovian sampling and show that the algorithm achieves a much smaller bias and variance error than the original Greedy-GQ. In particular, we prove that VR-Greedy-GQ achieves an improved sample complexity that is in the order of $\mathcal{O}(ε^{-2})$. We further compare the performance of VR-Greedy-GQ with that of Greedy-GQ in various RL experiments to corroborate our theoretical findings.

cs.LG↗

arXiv2021

Multi-Agent Off-Policy TD Learning: Finite-Time Analysis with Near-Optimal Sample Complexity and Communication Complexity

The finite-time convergence of off-policy TD learning has been comprehensively studied recently. However, such a type of convergence has not been well established for off-policy TD learning in the multi-agent setting, which covers broader applications and is fundamentally more challenging. This work develops two decentralized TD with correction (TDC) algorithms for multi-agent off-policy TD learning under Markovian sampling. In particular, our algorithms preserve full privacy of the actions, policies and rewards of the agents, and adopt mini-batch sampling to reduce the sampling variance and communication frequency. Under Markovian sampling and linear function approximation, we proved that the finite-time sample complexity of both algorithms for achieving an $ε$-accurate solution is in the order of $\mathcal{O}(ε^{-1}\ln ε^{-1})$, matching the near-optimal sample complexity of centralized TD(0) and TDC. Importantly, the communication complexity of our algorithms is in the order of $\mathcal{O}(\ln ε^{-1})$, which is significantly lower than the communication complexity $\mathcal{O}(ε^{-1}\ln ε^{-1})$ of the existing decentralized TD(0). Experiments corroborate our theoretical findings.

cs.LG↗

arXiv2021

Proximal Gradient Descent-Ascent: Variable Convergence under KŁ Geometry

The gradient descent-ascent (GDA) algorithm has been widely applied to solve minimax optimization problems. In order to achieve convergent policy parameters for minimax optimization, it is important that GDA generates convergent variable sequences rather than convergent sequences of function values or gradient norms. However, the variable convergence of GDA has been proved only under convexity geometries, and there lacks understanding for general nonconvex minimax optimization. This paper fills such a gap by studying the convergence of a more general proximal-GDA for regularized nonconvex-strongly-concave minimax optimization. Specifically, we show that proximal-GDA admits a novel Lyapunov function, which monotonically decreases in the minimax optimization process and drives the variable sequence to a critical point. By leveraging this Lyapunov function and the KŁ geometry that parameterizes the local geometries of general nonconvex functions, we formally establish the variable convergence of proximal-GDA to a critical point $x^*$, i.e., $x_t\to x^*, y_t\to y^*(x^*)$. Furthermore, over the full spectrum of the KŁ-parameterized geometry, we show that proximal-GDA achieves different types of convergence rates ranging from sublinear convergence up to finite-step convergence, depending on the geometry associated with the KŁ parameter. This is the first theoretical result on the variable convergence for nonconvex minimax optimization.

math.OC↗

arXiv2020

Neural Networked Assisted Tree Search for the Personnel Rostering Problem

The personnel rostering problem is the problem of finding an optimal way to assign employees to shifts, subject to a set of hard constraints which all valid solutions must follow, and a set of soft constraints which define the relative quality of valid solutions. The problem has received significant attention in the literature and is addressed by a large number of exact and metaheuristic methods. In order to make the complex and costly design of heuristics for the personnel rostering problem automatic, we propose a new method combined Deep Neural Network and Tree Search. By treating schedules as matrices, the neural network can predict the distance between the current solution and the optimal solution. It can select solution strategies by analyzing existing (near-)optimal solutions to personnel rostering problem instances. Combined with branch and bound, the network can give every node a probability which indicates the distance between it and the optimal one, so that a well-informed choice can be made on which branch to choose next and to prune the search tree.

cs.AI↗

arXiv2020

FedCluster: Boosting the Convergence of Federated Learning via Cluster-Cycling

We develop FedCluster--a novel federated learning framework with improved optimization efficiency, and investigate its theoretical convergence properties. The FedCluster groups the devices into multiple clusters that perform federated learning cyclically in each learning round. Therefore, each learning round of FedCluster consists of multiple cycles of meta-update that boost the overall convergence. In nonconvex optimization, we show that FedCluster with the devices implementing the local {stochastic gradient descent (SGD)} algorithm achieves a faster convergence rate than the conventional {federated averaging (FedAvg)} algorithm in the presence of device-level data heterogeneity. We conduct experiments on deep learning applications and demonstrate that FedCluster converges significantly faster than the conventional federated learning under diverse levels of device-level data heterogeneity for a variety of local optimizers.

cs.LG↗

arXiv2020

Momentum with Variance Reduction for Nonconvex Composition Optimization

Composition optimization is widely-applied in nonconvex machine learning. Various advanced stochastic algorithms that adopt momentum and variance reduction techniques have been developed for composition optimization. However, these algorithms do not fully exploit both techniques to accelerate the convergence and are lack of convergence guarantee in nonconvex optimization. This paper complements the existing literature by developing various momentum schemes with SPIDER-based variance reduction for non-convex composition optimization. In particular, our momentum design requires less number of proximal mapping evaluations per-iteration than that required by the existing Katyusha momentum. Furthermore, our algorithm achieves near-optimal sample complexity results in both non-convex finite-sum and online composition optimization and achieves a linear convergence rate under the gradient dominant condition. Numerical experiments demonstrate that our algorithm converges significantly faster than existing algorithms in nonconvex composition optimization.

cs.LG↗

arXiv2018

An LSTM Recurrent Network for Step Counting

Smartphones with sensors such as accelerometer and gyroscope can be used as pedometers and navigators. In this paper, we propose to use an LSTM recurrent network for counting the number of steps taken by both blind and sighted users, based on an annotated smartphone sensor dataset, WeAllWork. The models were trained separately for sighted people, blind people with a long cane or a guide dog for Leave-One-Out training modality. It achieved 5% overcount and undercount rate.

cs.LG↗