arXiv subjects

Songtao Lu

Publications and source records attributed to Songtao Lu.

At least 55 records · Page 3Linked to original sources

arXiv2022

Learning to Generate Image Source-Agnostic Universal Adversarial Perturbations

Adversarial perturbations are critical for certifying the robustness of deep learning models. A universal adversarial perturbation (UAP) can simultaneously attack multiple images, and thus offers a more unified threat model, obviating an image-wise attack algorithm. However, the existing UAP generator is underdeveloped when images are drawn from different image sources (e.g., with different image resolutions). Towards an authentic universality across image sources, we take a novel view of UAP generation as a customized instance of few-shot learning, which leverages bilevel optimization and learning-to-optimize (L2O) techniques for UAP generation with improved attack success rate (ASR). We begin by considering the popular model agnostic meta-learning (MAML) framework to meta-learn a UAP generator. However, we see that the MAML framework does not directly offer the universal attack across image sources, requiring us to integrate it with another meta-learning framework of L2O. The resulting scheme for meta-learning a UAP generator (i) has better performance (50% higher ASR) than baselines such as Projected Gradient Descent, (ii) has better performance (37% faster) than the vanilla L2O and MAML frameworks (when applicable), and (iii) is able to simultaneously handle UAP generation for different victim models and image data sources.

cs.LG↗

arXiv2022

Overcoming Catastrophic Forgetting via Direction-Constrained Optimization

This paper studies a new design of the optimization algorithm for training deep learning models with a fixed architecture of the classification network in a continual learning framework. The training data is non-stationary and the non-stationarity is imposed by a sequence of distinct tasks. We first analyze a deep model trained on only one learning task in isolation and identify a region in network parameter space, where the model performance is close to the recovered optimum. We provide empirical evidence that this region resembles a cone that expands along the convergence direction. We study the principal directions of the trajectory of the optimizer after convergence and show that traveling along a few top principal directions can quickly bring the parameters outside the cone but this is not the case for the remaining directions. We argue that catastrophic forgetting in a continual learning setting can be alleviated when the parameters are constrained to stay within the intersection of the plausible cones of individual tasks that were so far encountered during training. Based on this observation we present our direction-constrained optimization (DCO) method, where for each task we introduce a linear autoencoder to approximate its corresponding top forbidden principal directions. They are then incorporated into the loss function in the form of a regularization term for the purpose of learning the coming tasks without forgetting. Furthermore, in order to control the memory growth as the number of tasks increases, we propose a memory-efficient version of our algorithm called compressed DCO (DCO-COMP) that allocates a memory of fixed size for storing all autoencoders. We empirically demonstrate that our algorithm performs favorably compared to other state-of-art regularization-based continual learning methods.

cs.LG↗

arXiv2022

Understanding Latent Correlation-Based Multiview Learning and Self-Supervision: An Identifiability Perspective

Multiple views of data, both naturally acquired (e.g., image and audio) and artificially produced (e.g., via adding different noise to data samples), have proven useful in enhancing representation learning. Natural views are often handled by multiview analysis tools, e.g., (deep) canonical correlation analysis [(D)CCA], while the artificial ones are frequently used in self-supervised learning (SSL) paradigms, e.g., BYOL and Barlow Twins. Both types of approaches often involve learning neural feature extractors such that the embeddings of data exhibit high cross-view correlations. Although intuitive, the effectiveness of correlation-based neural embedding is mostly empirically validated. This work aims to understand latent correlation maximization-based deep multiview learning from a latent component identification viewpoint. An intuitive generative model of multiview data is adopted, where the views are different nonlinear mixtures of shared and private components. Since the shared components are view/distortion-invariant, representing the data using such components is believed to reveal the identity of the samples effectively and robustly. Under this model, latent correlation maximization is shown to guarantee the extraction of the shared components across views (up to certain ambiguities). In addition, it is further shown that the private information in each view can be provably disentangled from the shared using proper regularization design. A finite sample analysis, which has been rare in nonlinear mixture identifiability study, is also presented. The theoretical results and newly designed regularization are tested on a series of tasks.

cs.LG↗

arXiv2021

Zeroth-order Optimization for Composite Problems with Functional Constraints

In many real-world problems, first-order (FO) derivative evaluations are too expensive or even inaccessible. For solving these problems, zeroth-order (ZO) methods that only need function evaluations are often more efficient than FO methods or sometimes the only options. In this paper, we propose a novel zeroth-order inexact augmented Lagrangian method (ZO-iALM) to solve black-box optimization problems, which involve a composite (i.e., smooth+nonsmooth) objective and functional constraints. Under a certain regularity condition (also assumed by several existing works on FO methods), the query complexity of our ZO-iALM is $\tilde{O}(d\varepsilon^{-3})$ to find an $\varepsilon$-KKT point for problems with a nonconvex objective and nonconvex constraints, and $\tilde{O}(d\varepsilon^{-2.5})$ for nonconvex problems with convex constraints, where $d$ is the variable dimension. This appears to be the first work that develops an iALM-based ZO method for functional constrained optimization and meanwhile achieves query complexity results matching the best-known FO complexity results up to a factor of $d$. With an extensive experimental study, we show the effectiveness of our method. The applications of our method span from classical optimization problems to practical machine learning examples such as resource allocation in sensor networks and adversarial example generation.

math.OC↗

arXiv2021

Adversarial Examples can be Effective Data Augmentation for Unsupervised Machine Learning

Adversarial examples causing evasive predictions are widely used to evaluate and improve the robustness of machine learning models. However, current studies focus on supervised learning tasks, relying on the ground-truth data label, a targeted objective, or supervision from a trained classifier. In this paper, we propose a framework of generating adversarial examples for unsupervised models and demonstrate novel applications to data augmentation. Our framework exploits a mutual information neural estimator as an information-theoretic similarity measure to generate adversarial examples without supervision. We propose a new MinMax algorithm with provable convergence guarantees for efficient generation of unsupervised adversarial examples. Our framework can also be extended to supervised adversarial examples. When using unsupervised adversarial examples as a simple plug-in data augmentation tool for model retraining, significant improvements are consistently observed across different unsupervised tasks and datasets, including data reconstruction, representation learning, and contrastive learning. Our results show novel methods and considerable advantages in studying and improving unsupervised machine learning via adversarial examples.

cs.LG↗

arXiv2021

Optimal Discrete Constellation Inputs for Aggregated LiFi-WiFi Networks

In this paper, we investigate the performance of a practical aggregated LiFi-WiFi system with the discrete constellation inputs from a practical view. We derive the achievable rate expressions of the aggregated LiFi-WiFi system for the first time. Then, we study the rate maximization problem via optimizing the constellation distribution and power allocation jointly. Specifically, a multilevel mercy-filling power allocation scheme is proposed by exploiting the relationship between the mutual information and minimum mean-squared error (MMSE) of discrete inputs. Meanwhile, an inexact gradient descent method is proposed for obtaining the optimal probability distributions. To strike a balance between the computational complexity and the transmission performance, we further develop a framework that maximizes the lower bound of the achievable rate where the optimal power allocation can be obtained in closed forms and the constellation distributions problem can be solved efficiently by Frank-Wolfe method. Extensive numerical results show that the optimized strategies are able to provide significant gains over the state-of-the-art schemes in terms of the achievable rate.

cs.IT↗

arXiv2021

Signal Transformer: Complex-valued Attention and Meta-Learning for Signal Recognition

Deep neural networks have been shown as a class of useful tools for addressing signal recognition issues in recent years, especially for identifying the nonlinear feature structures of signals. However, this power of most deep learning techniques heavily relies on an abundant amount of training data, so the performance of classic neural nets decreases sharply when the number of training data samples is small or unseen data are presented in the testing phase. This calls for an advanced strategy, i.e., model-agnostic meta-learning (MAML), which is able to capture the invariant representation of the data samples or signals. In this paper, inspired by the special structure of the signal, i.e., real and imaginary parts consisted in practical time-series signals, we propose a Complex-valued Attentional MEta Learner (CAMEL) for the problem of few-shot signal recognition by leveraging attention and meta-learning in the complex domain. To the best of our knowledge, this is also the first complex-valued MAML that can find the first-order stationary points of general nonconvex problems with theoretical convergence guarantees. Extensive experiments results showcase the superiority of the proposed CAMEL compared with the state-of-the-art methods.

cs.LG↗

arXiv2021

An Efficient Learning Framework For Federated XGBoost Using Secret Sharing And Distributed Optimization

XGBoost is one of the most widely used machine learning models in the industry due to its superior learning accuracy and efficiency. Targeting at data isolation issues in the big data problems, it is crucial to deploy a secure and efficient federated XGBoost (FedXGB) model. Existing FedXGB models either have data leakage issues or are only applicable to the two-party setting with heavy communication and computation overheads. In this paper, a lossless multi-party federated XGB learning framework is proposed with a security guarantee, which reshapes the XGBoost's split criterion calculation process under a secret sharing setting and solves the leaf weight calculation problem by leveraging distributed optimization. Remarkably, a thorough analysis of model security is provided as well, and multiple numerical results showcase the superiority of the proposed FedXGB compared with the state-of-the-art models on benchmark datasets.

cs.LG↗

arXiv2021

ScaleCom: Scalable Sparsified Gradient Compression for Communication-Efficient Distributed Training

Large-scale distributed training of Deep Neural Networks (DNNs) on state-of-the-art platforms is expected to be severely communication constrained. To overcome this limitation, numerous gradient compression techniques have been proposed and have demonstrated high compression ratios. However, most existing methods do not scale well to large scale distributed systems (due to gradient build-up) and/or fail to evaluate model fidelity (test accuracy) on large datasets. To mitigate these issues, we propose a new compression technique, Scalable Sparsified Gradient Compression (ScaleCom), that leverages similarity in the gradient distribution amongst learners to provide significantly improved scalability. Using theoretical analysis, we show that ScaleCom provides favorable convergence guarantees and is compatible with gradient all-reduce techniques. Furthermore, we experimentally demonstrate that ScaleCom has small overheads, directly reduces gradient traffic and provides high compression rates (65-400X) and excellent scalability (up to 64 learners and 8-12X larger batch sizes over standard training) across a wide range of applications (image, language, and speech) without significant accuracy loss.

cs.LG↗

arXiv2021

Robust Beamforming Design for Covert Communications

In this paper, we consider a common unicast beamforming network where Alice utilizes the communication to Carol as a cover and covertly transmits a message to Bob without being recognized by Willie. We investigate the beamformer design of Alice to maximize the covert rate to Bob when Alice has either perfect or imperfect knowledge about Willie's channel state information (WCSI). For the perfect WCSI case, the problem is formulated under the perfect covert constraint, and we develop a covert beamformer by applying semidefinite relaxation and the bisection method. Then, to reduce the computational complexity, we further propose a zero-forcing beamformer design with a single iteration processing. For the case of the imperfect WCSI, the robust beamformer is developed based on a relaxation and restriction approach by utilizing the property of Kullback-Leibler divergence. Furthermore, we derive the optimal decision threshold of Willie, and analyze the false alarm and the missed detection probabilities in this case. Finally, the performance of the proposed beamformer designs is evaluated through numerical experiments.

cs.IT↗

arXiv2021

Rate-improved Inexact Augmented Lagrangian Method for Constrained Nonconvex Optimization

First-order methods have been studied for nonlinear constrained optimization within the framework of the augmented Lagrangian method (ALM) or penalty method. We propose an improved inexact ALM (iALM) and conduct a unified analysis for nonconvex problems with either affine equality or nonconvex constraints. Under certain regularity conditions (that are also assumed by existing works), we show an $\tilde{O}(\varepsilon^{-\frac{5}{2}})$ complexity result for a problem with a nonconvex objective and affine equality constraints and an $\tilde{O}(\varepsilon^{-3})$ complexity result for a problem with a nonconvex objective and nonconvex constraints, where the complexity is measured by the number of first-order oracles to yield an $\varepsilon$-KKT solution. Both results are the best known. The same-order complexity results have been achieved by penalty methods. However, two different analysis techniques are used to obtain the results, and more importantly, the penalty methods generally perform significantly worse than iALM in practice. Our improved iALM and analysis close the gap between theory and practice. Numerical experiments on nonconvex problems with affine equality or nonconvex constraints are provided to demonstrate the effectiveness of our proposed method.

math.OC↗

arXiv2021

Hybrid Block Successive Approximation for One-Sided Non-Convex Min-Max Problems: Algorithms and Applications

The min-max problem, also known as the saddle point problem, is a class of optimization problems which minimizes and maximizes two subsets of variables simultaneously. This class of problems can be used to formulate a wide range of signal processing and communication (SPCOM) problems. Despite its popularity, most existing theory for this class has been mainly developed for problems with certain special convex-concave structure. Therefore, it cannot be used to guide the algorithm design for many interesting problems in SPCOM, where various kinds of non-convexity arise. In this work, we consider a block-wise one-sided non-convex min-max problem, in which the minimization problem consists of multiple blocks and is non-convex, while the maximization problem is (strongly) concave. We propose a class of simple algorithms named Hybrid Block Successive Approximation (HiBSA), which alternatingly perform gradient descent-type steps for the minimization blocks and gradient ascent-type steps for the maximization problem. A key element in the proposed algorithm is the use of certain regularization and penalty sequences, which stabilize the algorithm and ensure convergence. We show that HiBSA converges to some properly defined first-order stationary solutions with quantifiable global rates. To validate the efficiency of the proposed algorithms, we conduct numerical tests on a number of problems, including the robust learning problem, the non-convex min-utility maximization problems, and certain wireless jamming problem arising in interfering channels.

math.OC↗

arXiv2021

Federated Acoustic Modeling For Automatic Speech Recognition

Data privacy and protection is a crucial issue for any automatic speech recognition (ASR) service provider when dealing with clients. In this paper, we investigate federated acoustic modeling using data from multiple clients. A client's data is stored on a local data server and the clients communicate only model parameters with a central server, and not their data. The communication happens infrequently to reduce the communication cost. To mitigate the non-iid issue, client adaptive federated training (CAFT) is proposed to canonicalize data across clients. The experiments are carried out on 1,150 hours of speech data from multiple domains. Hybrid LSTM acoustic models are trained via federated learning and their performance is compared to traditional centralized acoustic model training. The experimental results demonstrate the effectiveness of the proposed federated acoustic modeling strategy. We also show that CAFT can further improve the performance of the federated acoustic model.

cs.SD↗

arXiv2020

Non-convex Min-Max Optimization: Applications, Challenges, and Recent Theoretical Advances

The min-max optimization problem, also known as the saddle point problem, is a classical optimization problem which is also studied in the context of zero-sum games. Given a class of objective functions, the goal is to find a value for the argument which leads to a small objective value even for the worst case function in the given class. Min-max optimization problems have recently become very popular in a wide range of signal and data processing applications such as fair beamforming, training generative adversarial networks (GANs), and robust machine learning, to just name a few. The overarching goal of this article is to provide a survey of recent advances for an important subclass of min-max problem, where the minimization and maximization problems can be non-convex and/or non-concave. In particular, we will first present a number of applications to showcase the importance of such min-max problems; then we discuss key theoretical challenges, and provide a selective review of some exciting recent theoretical and algorithmic advances in tackling non-convex min-max problems. Finally, we will point out open questions and future research directions.

math.OC↗

arXiv2020

Min-Max Optimization without Gradients: Convergence and Applications to Adversarial ML

In this paper, we study the problem of constrained robust (min-max) optimization ina black-box setting, where the desired optimizer cannot access the gradients of the objective function but may query its values. We present a principled optimization framework, integrating a zeroth-order (ZO) gradient estimator with an alternating projected stochastic gradient descent-ascent method, where the former only requires a small number of function queries and the later needs just one-step descent/ascent update. We show that the proposed framework, referred to as ZO-Min-Max, has a sub-linear convergence rate under mild conditions and scales gracefully with problem size. From an application side, we explore a promising connection between black-box min-max optimization and black-box evasion and poisoning attacks in adversarial machine learning (ML). Our empirical evaluations on these use cases demonstrate the effectiveness of our approach and its scalability to dimensions that prohibit using recent black-box solvers.

cs.LG↗

arXiv2020

No-regret Non-convex Online Meta-Learning

The online meta-learning framework is designed for the continual lifelong learning setting. It bridges two fields: meta-learning which tries to extract prior knowledge from past tasks for fast learning of future tasks, and online-learning which deals with the sequential setting where problems are revealed one by one. In this paper, we generalize the original framework from convex to non-convex setting, and introduce the local regret as the alternative performance measure. We then apply this framework to stochastic settings, and show theoretically that it enjoys a logarithmic local regret, and is robust to any hyperparameter initialization. The empirical test on a real-world task demonstrates its superiority compared with traditional methods.

cs.LG↗

arXiv2020

Randomized Bregman Coordinate Descent Methods for Non-Lipschitz Optimization

We propose a new \textit{randomized Bregman (block) coordinate descent} (RBCD) method for minimizing a composite problem, where the objective function could be either convex or nonconvex, and the smooth part are freed from the global Lipschitz-continuous (partial) gradient assumption. Under the notion of relative smoothness based on the Bregman distance, we prove that every limit point of the generated sequence is a stationary point. Further, we show that the iteration complexity of the proposed method is $O(n\varepsilon^{-2})$ to achieve $ε$-stationary point, where $n$ is the number of blocks of coordinates. If the objective is assumed to be convex, the iteration complexity is improved to $O(nε^{-1} )$. If, in addition, the objective is strongly convex (relative to the reference function), the global linear convergence rate is recovered. We also present the accelerated version of the RBCD method, which attains an $O(n\varepsilon^{-1/γ} )$ iteration complexity for the convex case, where the scalar $γ\in [1,2]$ is determined by the \textit{generalized translation variant} of the Bregman distance. Convergence analysis without assuming the global Lipschitz-continuous (partial) gradient sets our results apart from the existing works in the composite problems.

math.OC↗

arXiv2020

Distributed Learning in the Non-Convex World: From Batch to Streaming Data, and Beyond

Distributed learning has become a critical enabler of the massively connected world envisioned by many. This article discusses four key elements of scalable distributed processing and real-time intelligence --- problems, data, communication and computation. Our aim is to provide a fresh and unique perspective about how these elements should work together in an effective and coherent manner. In particular, we {provide a selective review} about the recent techniques developed for optimizing non-convex models (i.e., problem classes), processing batch and streaming data (i.e., data types), over the networks in a distributed manner (i.e., communication and computation paradigm). We describe the intuitions and connections behind a core set of popular distributed algorithms, emphasizing how to trade off between computation and communication costs. Practical issues and future research directions will also be discussed.

cs.LG↗