arXiv subjects

Chenchen Liu

Publications and source records attributed to Chenchen Liu.

34 records · Page 2Linked to original sources

arXiv2023

GACER: Granularity-Aware ConcurrEncy Regulation for Multi-Tenant Deep Learning

As deep learning continues to advance and is applied to increasingly complex scenarios, the demand for concurrent deployment of multiple neural network models has arisen. This demand, commonly referred to as multi-tenant computing, is becoming more and more important. However, even the most mature GPU-based computing systems struggle to adequately address the significant heterogeneity and complexity among concurrent models in terms of resource allocation and runtime scheduling. And this usually results in considerable resource utilization and throughput issues. To tackle these issues, this work proposes a set of optimization techniques that advance the granularity of computing management from both the spatial and temporal perspectives, specifically tailored to heterogeneous model compositions for deep learning inference and training. These techniques are further integrated as GACER -- an automated optimization framework that provides high-utilization, high-throughput, and low-latency multi-tenant computing support. And our experiments demonstrate that GACER significantly improves the overall resource utilization and consistently achieves outstanding speedups compared to native GPU computing frameworks and existing state-of-the-art optimization works.

cs.DC↗

arXiv2022

A Survey of Multi-Tenant Deep Learning Inference on GPU

Deep Learning (DL) models have achieved superior performance. Meanwhile, computing hardware like NVIDIA GPUs also demonstrated strong computing scaling trends with 2x throughput and memory bandwidth for each generation. With such strong computing scaling of GPUs, multi-tenant deep learning inference by co-locating multiple DL models onto the same GPU becomes widely deployed to improve resource utilization, enhance serving throughput, reduce energy cost, etc. However, achieving efficient multi-tenant DL inference is challenging which requires thorough full-stack system optimization. This survey aims to summarize and categorize the emerging challenges and optimization opportunities for multi-tenant DL inference on GPU. By overviewing the entire optimization stack, summarizing the multi-tenant computing innovations, and elaborating the recent technological advances, we hope that this survey could shed light on new optimization perspectives and motivate novel works in future large-scale DL system optimization.

cs.DC↗

arXiv2022

Heterogeneous Federated Learning

Federated learning learns from scattered data by fusing collaborative models from local nodes. However, due to chaotic information distribution, the model fusion may suffer from structural misalignment with regard to unmatched parameters. In this work, we propose a novel federated learning framework to resolve this issue by establishing a firm structure-information alignment across collaborative models. Specifically, we design a feature-oriented regulation method ({$Ψ$-Net}) to ensure explicit feature information allocation in different neural network structures. Applying this regulating method to collaborative models, matchable structures with similar feature information can be initialized at the very early training stage. During the federated learning process under either IID or non-IID scenarios, dedicated collaboration schemes further guarantee ordered information distribution with definite structure matching, so as the comprehensive model alignment. Eventually, this framework effectively enhances the federated learning applicability to extensive heterogeneous settings, while providing excellent convergence speed, accuracy, and computation/communication efficiency.

cs.LG↗

arXiv2022

A Survey of Large-Scale Deep Learning Serving System Optimization: Challenges and Opportunities

Deep Learning (DL) models have achieved superior performance in many application domains, including vision, language, medical, commercial ads, entertainment, etc. With the fast development, both DL applications and the underlying serving hardware have demonstrated strong scaling trends, i.e., Model Scaling and Compute Scaling, for example, the recent pre-trained model with hundreds of billions of parameters with ~TB level memory consumption, as well as the newest GPU accelerators providing hundreds of TFLOPS. With both scaling trends, new problems and challenges emerge in DL inference serving systems, which gradually trends towards Large-scale Deep learning Serving systems (LDS). This survey aims to summarize and categorize the emerging challenges and optimization opportunities for large-scale deep learning serving systems. By providing a novel taxonomy, summarizing the computing paradigms, and elaborating the recent technique advances, we hope that this survey could shed light on new optimization perspectives and motivate novel works in large-scale deep learning system optimization.

cs.LG↗

arXiv2021

Fed2: Feature-Aligned Federated Learning

Federated learning learns from scattered data by fusing collaborative models from local nodes. However, the conventional coordinate-based model averaging by FedAvg ignored the random information encoded per parameter and may suffer from structural feature misalignment. In this work, we propose Fed2, a feature-aligned federated learning framework to resolve this issue by establishing a firm structure-feature alignment across the collaborative models. Fed2 is composed of two major designs: First, we design a feature-oriented model structure adaptation method to ensure explicit feature allocation in different neural network structures. Applying the structure adaptation to collaborative models, matchable structures with similar feature information can be initialized at the very early training stage. During the federated learning process, we then propose a feature paired averaging scheme to guarantee aligned feature distribution and maintain no feature fusion conflicts under either IID or non-IID scenarios. Eventually, Fed2 could effectively enhance the federated learning convergence performance under extensive homo- and heterogeneous settings, providing excellent convergence speed, accuracy, and computation/communication efficiency.

cs.LG↗

arXiv2021

Automated Runtime-Aware Scheduling for Multi-Tenant DNN Inference on GPU

With the fast development of deep neural networks (DNNs), many real-world applications are adopting multiple models to conduct compound tasks, such as co-running classification, detection, and segmentation models on autonomous vehicles. Such multi-tenant DNN inference cases greatly exacerbate the computational complexity and call for comprehensive collaboration for graph-level operator scheduling, runtime-level resource awareness, as well as hardware scheduler support. However, the current scheduling support for such multi-tenant inference is still relatively backward. In this work, we propose a resource-aware scheduling framework for efficient multi-tenant DNN inference on GPU, which automatically coordinates DNN computing in different execution levels. Leveraging the unified scheduling intermediate representation and the automated ML-based searching algorithm, optimal schedules could be generated to wisely adjust model concurrency and interleave DNN model operators, maintaining a continuously balanced resource utilization across the entire inference process, and eventually improving the runtime efficiency. Experiments show that we could consistently achieve 1.3-1.7x speed-up, compared to regular DNN runtime libraries (e.g., CuDNN, TVM) and particular concurrent scheduling methods (e.g., NVIDIA Multi-Stream).

cs.DC↗

arXiv2021

A Survey on Fundamental Limits of Integrated Sensing and Communication

The integrated sensing and communication (ISAC), in which the sensing and communication share the same frequency band and hardware, has emerged as a key technology in future wireless systems. Early works on ISAC have been focused on the design, analysis and optimization of practical ISAC technologies for various ISAC systems. While this line of works are necessary, it is equally important to study the fundamental limits of ISAC in order to understand the gap between the current state-of-the-art technologies and the performance limits, and provide useful insights and guidance for the development of better ISAC technologies that can approach the performance limits. In this paper, we aim to provide a comprehensive survey for the current research progress on the fundamental limits of ISAC. Particularly, we first propose a systematic classification method for both traditional radio sensing (such as radar sensing and wireless localization) and ISAC so that they can be naturally incorporated into a unified framework. Then we summarize the major performance metrics and bounds used in sensing, communications and ISAC, respectively. After that, we present the current research progresses on fundamental limits of each class of the traditional sensing and ISAC systems. Finally, the open problems and future research directions are discussed.

cs.IT↗

arXiv2020

AntiDote: Attention-based Dynamic Optimization for Neural Network Runtime Efficiency

Convolutional Neural Networks (CNNs) achieved great cognitive performance at the expense of considerable computation load. To relieve the computation load, many optimization works are developed to reduce the model redundancy by identifying and removing insignificant model components, such as weight sparsity and filter pruning. However, these works only evaluate model components' static significance with internal parameter information, ignoring their dynamic interaction with external inputs. With per-input feature activation, the model component significance can dynamically change, and thus the static methods can only achieve sub-optimal results. Therefore, we propose a dynamic CNN optimization framework in this work. Based on the neural network attention mechanism, we propose a comprehensive dynamic optimization framework including (1) testing-phase channel and column feature map pruning, as well as (2) training-phase optimization by targeted dropout. Such a dynamic optimization framework has several benefits: (1) First, it can accurately identify and aggressively remove per-input feature redundancy with considering the model-input interaction; (2) Meanwhile, it can maximally remove the feature map redundancy in various dimensions thanks to the multi-dimension flexibility; (3) The training-testing co-optimization favors the dynamic pruning and helps maintain the model accuracy even with very high feature pruning ratio. Extensive experiments show that our method could bring 37.4% to 54.5% FLOPs reduction with negligible accuracy drop on various of test networks.

cs.CV↗

arXiv2019

Functionality-Oriented Convolutional Filter Pruning

The sophisticated structure of Convolutional Neural Network (CNN) allows for outstanding performance, but at the cost of intensive computation. As significant redundancies inevitably present in such a structure, many works have been proposed to prune the convolutional filters for computation cost reduction. Although extremely effective, most works are based only on quantitative characteristics of the convolutional filters, and highly overlook the qualitative interpretation of individual filter's specific functionality. In this work, we interpreted the functionality and redundancy of the convolutional filters from different perspectives, and proposed a functionality-oriented filter pruning method. With extensive experiment results, we proved the convolutional filters' qualitative significance regardless of magnitude, demonstrated significant neural network redundancy due to repetitive filter functions, and analyzed the filter functionality defection under inappropriate retraining process. Such an interpretable pruning approach not only offers outstanding computation cost optimization over previous filter pruning methods, but also interprets filter pruning process.

cs.LG↗

arXiv2019

Interpreting and Evaluating Neural Network Robustness

Recently, adversarial deception becomes one of the most considerable threats to deep neural networks. However, compared to extensive research in new designs of various adversarial attacks and defenses, the neural networks' intrinsic robustness property is still lack of thorough investigation. This work aims to qualitatively interpret the adversarial attack and defense mechanism through loss visualization, and establish a quantitative metric to evaluate the neural network model's intrinsic robustness. The proposed robustness metric identifies the upper bound of a model's prediction divergence in the given domain and thus indicates whether the model can maintain a stable prediction. With extensive experiments, our metric demonstrates several advantages over conventional adversarial testing accuracy based robustness estimation: (1) it provides a uniformed evaluation to models with different structures and parameter scales; (2) it over-performs conventional accuracy based robustness estimation and provides a more reliable evaluation that is invariant to different test settings; (3) it can be fast generated without considerable testing cost.

cs.LG↗

arXiv2018

Interpreting Adversarial Robustness: A View from Decision Surface in Input Space

One popular hypothesis of neural network generalization is that the flat local minima of loss surface in parameter space leads to good generalization. However, we demonstrate that loss surface in parameter space has no obvious relationship with generalization, especially under adversarial settings. Through visualizing decision surfaces in both parameter space and input space, we instead show that the geometry property of decision surface in input space correlates well with the adversarial robustness. We then propose an adversarial robustness indicator, which can evaluate a neural network's intrinsic robustness property without testing its accuracy under adversarial attacks. Guided by it, we further propose our robust training method. Without involving adversarial training, our method could enhance network's intrinsic adversarial robustness against various adversarial attacks.

cs.LG↗

arXiv2018

HASP: A High-Performance Adaptive Mobile Security Enhancement Against Malicious Speech Recognition

Nowadays, machine learning based Automatic Speech Recognition (ASR) technique has widely spread in smartphones, home devices, and public facilities. As convenient as this technology can be, a considerable security issue also raises -- the users' speech content might be exposed to malicious ASR monitoring and cause severe privacy leakage. In this work, we propose HASP -- a high-performance security enhancement approach to solve this security issue on mobile devices. Leveraging ASR systems' vulnerability to the adversarial examples, HASP is designed to cast human imperceptible adversarial noises to real-time speech and effectively perturb malicious ASR monitoring by increasing the Word Error Rate (WER). To enhance the practical performance on mobile devices, HASP is also optimized for effective adaptation to the human speech characteristics, environmental noises, and mobile computation scenarios. The experiments show that HASP can achieve optimal real-time security enhancement: it can lead an average WER of 84.55% for perturbing the malicious ASR monitoring, and the data processing speed is 15x to 40x faster compared to the state-of-the-art methods. Moreover, HASP can effectively perturb various ASR systems, demonstrating a strong transferability.

cs.CR↗

arXiv2018

Towards Robust Training of Neural Networks by Regularizing Adversarial Gradients

In recent years, neural networks have demonstrated outstanding effectiveness in a large amount of applications.However, recent works have shown that neural networks are susceptible to adversarial examples, indicating possible flaws intrinsic to the network structures. To address this problem and improve the robustness of neural networks, we investigate the fundamental mechanisms behind adversarial examples and propose a novel robust training method via regulating adversarial gradients. The regulation effectively squeezes the adversarial gradients of neural networks and significantly increases the difficulty of adversarial example generation.Without any adversarial example involved, the robust training method could generate naturally robust networks, which are near-immune to various types of adversarial examples. Experiments show the naturally robust networks can achieve optimal accuracy against Fast Gradient Sign Method (FGSM) and C\&W attacks on MNIST, Cifar10, and Google Speech Command dataset. Moreover, our proposed method also provides neural networks with consistent robustness against transferable attacks.

cs.LG↗

arXiv2018

How convolutional neural network see the world - A survey of convolutional neural network visualization methods

Nowadays, the Convolutional Neural Networks (CNNs) have achieved impressive performance on many computer vision related tasks, such as object detection, image recognition, image retrieval, etc. These achievements benefit from the CNNs outstanding capability to learn the input features with deep layers of neuron structures and iterative training process. However, these learned features are hard to identify and interpret from a human vision perspective, causing a lack of understanding of the CNNs internal working mechanism. To improve the CNN interpretability, the CNN visualization is well utilized as a qualitative analysis method, which translates the internal features into visually perceptible patterns. And many CNN visualization works have been proposed in the literature to interpret the CNN in perspectives of network structure, operation, and semantic concept. In this paper, we expect to provide a comprehensive survey of several representative CNN visualization methods, including Activation Maximization, Network Inversion, Deconvolutional Neural Networks (DeconvNet), and Network Dissection based visualization. These methods are presented in terms of motivations, algorithms, and experiment results. Based on these visualization methods, we also discuss their practical applications to demonstrate the significance of the CNN interpretability in areas of network design, optimization, security enhancement, etc.

cs.CV↗

arXiv2015

Computational homogenization of heterogeneous media under dynamic loading

A variational coarse-graining framework for heterogeneous media is developed that allows for a seamless transition from the traditional static scenario to a arbitrary loading conditions, including inertia effects and body forces. The strategy is formulated in the spirit of computational homogenization methods (FE$^2$) and is based on the discrete version of Hill's averaging results recently derived by the authors. In particular, the traditional static multiscale scheme is proved here to be equivalent to a direct homogenization of the principle of minimum potential energy and to hold exactly under a finite element discretization. This perspective provides a unifying variational framework for the FE$^2$ method, in the static setting, with Dirichlet or Neumann boundary conditions on the representative volume element; and it directly manifests the approximate duality of the effective strain energy density obtained with these two types of boundary conditions in the sense of Legendre transformation. Its generalization to arbitrary loading conditions and material constitutive relations is then immediate through the incremental minimum formulation of the dynamic problem à la Radovitzky and Ortiz (1999), which, in the discrete setting, is in full analogy to the static problem. These theoretical developments are then translated into an efficient multiscale FE$^2$ computational strategy for the homogenization of a microscopic explicit dynamics scheme, with two noteworthy properties. Firstly, each time incremental problem can be solved exactly with a single Newton-Raphson iteration with a constant Hessian, regardless of the specific non-linearities or history-dependence of the micro-constituents' behavior. Secondly, the scheme concurrently solves for the microscopic and macroscopic degrees of freedom, in contrast to standard approaches based on sequential or nested minimizations.

cond-mat.mtrl-sci↗

arXiv2015

Discrete averaging relations for micro to macro transition

The well-known Hill's averaging theorems for stresses and strains as well as the so-called Hill-Mandel principle of macrohomogeneity are essential ingredients for the coupling and the consistency between the micro and macro scales in multiscale finite element procedures (FE$^2$). We show in this paper that these averaging relations hold exactly under standard finite element discretizations, even if the stress field is discontinuous across elements and the standard proofs based on the divergence theorem are no longer suitable. The discrete averaging results are derived for the three classical types of boundary conditions (affine displacement, periodic and uniform traction boundary conditions) using the properties of the shape functions and the weak form of the microscopic equilibrium equations. The analytical proofs are further verified numerically through a simple finite element simulation of an irregular representative volume element undergoing large deformations. Furthermore, the proofs are extended to include the effects of body forces and inertia, and the results are consistent with those in the smooth continuum setting. This work provides a solid foundation to apply Hill's averaging relations in multiscale finite element methods without introducing an additional error in the scale transition due to the discretization.

cond-mat.mtrl-sci↗