arXiv subjects

Saurabh Agarwal

Publications and source records attributed to Saurabh Agarwal.

27 records · Page 2Linked to original sources

arXiv2021

On the Utility of Gradient Compression in Distributed Training Systems

A rich body of prior work has highlighted the existence of communication bottlenecks in synchronous data-parallel training. To alleviate these bottlenecks, a long line of recent work proposes gradient and model compression methods. In this work, we evaluate the efficacy of gradient compression methods and compare their scalability with optimized implementations of synchronous data-parallel SGD across more than 200 different setups. Surprisingly, we observe that only in 6 cases out of more than 200, gradient compression methods provide speedup over optimized synchronous data-parallel training in the typical data-center setting. We conduct an extensive investigation to identify the root causes of this phenomenon, and offer a performance model that can be used to identify the benefits of gradient compression for a variety of system setups. Based on our analysis, we propose a list of desirable properties that gradient compression methods should satisfy, in order for them to provide a meaningful end-to-end speedup.

cs.DC↗

arXiv2021

AutoFreeze: Automatically Freezing Model Blocks to Accelerate Fine-tuning

With the rapid adoption of machine learning (ML), a number of domains now use the approach of fine tuning models which were pre-trained on a large corpus of data. However, our experiments show that even fine-tuning on models like BERT can take many hours even when using modern accelerators like GPUs. While prior work proposes limiting the number of layers that are fine-tuned, e.g., freezing all layers but the last layer, we find that such static approaches lead to reduced accuracy. We propose, AutoFreeze, a system that uses an adaptive approach to choose which layers are trained and show how this can accelerate model fine-tuning while preserving accuracy. We also develop mechanisms to enable efficient caching of intermediate activations which can reduce the forward computation time when performing fine-tuning. We extend AutoFreeze to perform distributed fine-tuning and design two execution modes that minimize cost and running time respectively. Our evaluation on ten NLP tasks shows that AutoFreeze, with caching enabled, can improve fine-tuning on a single GPU by up to 2.55x. On a 64 GPU cluster, for fine-tuning on the AG's news dataset, AutoFreeze is able to achieve up to 4.38x speedup when optimizing for end-to-end training time and 5.03x reduction in total cost when optimizing for efficiency, without affecting model accuracy.

cs.LG↗

arXiv2021

Pufferfish: Communication-efficient Models At No Extra Cost

To mitigate communication overheads in distributed model training, several studies propose the use of compressed stochastic gradients, usually achieved by sparsification or quantization. Such techniques achieve high compression ratios, but in many cases incur either significant computational overheads or some accuracy loss. In this work, we present Pufferfish, a communication and computation efficient distributed training framework that incorporates the gradient compression into the model training process via training low-rank, pre-factorized deep networks. Pufferfish not only reduces communication, but also completely bypasses any computation overheads related to compression, and achieves the same accuracy as state-of-the-art, off-the-shelf deep models. Pufferfish can be directly integrated into current deep learning frameworks with minimum implementation modification. Our extensive experiments over real distributed setups, across a variety of large-scale machine learning tasks, indicate that Pufferfish achieves up to 1.64x end-to-end speedup over the latest distributed training API in PyTorch without accuracy loss. Compared to the Lottery Ticket Hypothesis models, Pufferfish leads to equally accurate, small-parameter models while avoiding the burden of "winning the lottery". Pufferfish also leads to more accurate and smaller models than SOTA structured model pruning methods.

cs.LG↗

arXiv2020

Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification

Distributed model training suffers from communication bottlenecks due to frequent model updates transmitted across compute nodes. To alleviate these bottlenecks, practitioners use gradient compression techniques like sparsification, quantization, or low-rank updates. The techniques usually require choosing a static compression ratio, often requiring users to balance the trade-off between model accuracy and per-iteration speedup. In this work, we show that such performance degradation due to choosing a high compression ratio is not fundamental. An adaptive compression strategy can reduce communication while maintaining final test accuracy. Inspired by recent findings on critical learning regimes, in which small gradient errors can have irrecoverable impact on model performance, we propose Accordion a simple yet effective adaptive compression algorithm. While Accordion maintains a high enough compression rate on average, it avoids over-compressing gradients whenever in critical learning regimes, detected by a simple gradient-norm based criterion. Our extensive experimental study over a number of machine learning tasks in distributed environments indicates that Accordion, maintains similar model accuracy to uncompressed training, yet achieves up to 5.5x better compression and up to 4.1x end-to-end speedup over static approaches. We show that Accordion also works for adjusting the batch size, another popular strategy for alleviating communication bottlenecks.

cs.LG↗

arXiv2020

Attack of the Tails: Yes, You Really Can Backdoor Federated Learning

Due to its decentralized nature, Federated Learning (FL) lends itself to adversarial attacks in the form of backdoors during training. The goal of a backdoor is to corrupt the performance of the trained model on specific sub-tasks (e.g., by classifying green cars as frogs). A range of FL backdoor attacks have been introduced in the literature, but also methods to defend against them, and it is currently an open question whether FL systems can be tailored to be robust against backdoors. In this work, we provide evidence to the contrary. We first establish that, in the general case, robustness to backdoors implies model robustness to adversarial examples, a major open problem in itself. Furthermore, detecting the presence of a backdoor in a FL model is unlikely assuming first order oracles or polynomial time. We couple our theoretical results with a new family of backdoor attacks, which we refer to as edge-case backdoors. An edge-case backdoor forces a model to misclassify on seemingly easy inputs that are however unlikely to be part of the training, or test data, i.e., they live on the tail of the input distribution. We explain how these edge-case backdoors can lead to unsavory failures and may have serious repercussions on fairness, and exhibit that with careful tuning at the side of the adversary, one can insert them across a range of machine learning tasks (e.g., image classification, OCR, text prediction, sentiment analysis).

cs.LG↗

arXiv2019

Scalable K-Medoids via True Error Bound and Familywise Bandits

K-Medoids(KM) is a standard clustering method, used extensively on semi-metric data.Error analyses of KM have traditionally used an in-sample notion of error,which can be far from the true error and suffer from generalization gap. We formalize the true K-Medoid error based on the underlying data distribution.We decompose the true error into fundamental statistical problems of: minimum estimation (ME) and minimum mean estimation (MME). We provide a convergence result for MME. We show $\errMME$ decreases no slower than $Θ(\frac{1}{n^{\frac{2}{3}}})$, where $n$ is a measure of sample size. Inspired by this bound, we propose a computationally efficient, distributed KM algorithm namely MCPAM. MCPAM has expected runtime $\mathcal{O}(km)$,where $k$ is the number of medoids and $m$ is number of samples. MCPAM provides massive computational savings for a small tradeoff in accuracy. We verify the quality and scaling properties of MCPAM on various datasets. And achieve the hitherto unachieved feat of calculating the KM of 1 billion points on semi-metric spaces.

cs.LG↗

arXiv2018

Graph based Question Answering System

In today's digital age in the dawning era of big data analytics it is not the information but the linking of information through entities and actions which defines the discourse. Any textual data either available on the Internet off off-line (like newspaper data, Wikipedia dump, etc) is basically connect information which cannot be treated isolated for its wholesome semantics. There is a need for an automated retrieval process with proper information extraction to structure the data for relevant and fast text analytics. The first big challenge is the conversion of unstructured textual data to structured data. Unlike other databases, graph databases handle relationships and connections elegantly. Our project aims at developing a graph-based information extraction and retrieval system.

cs.CL↗

arXiv2015

Framework for Application Mapping over Packet-Switched Network of FPGAs: Case Studies

The algorithm-to-hardware High-level synthesis (HLS) tools today are purported to produce hardware comparable in quality to handcrafted designs, particularly with user directive driven or domains specific HLS. However, HLS tools are not readily equipped for when an application/algorithm needs to scale. We present a (work-in-progress) semi-automated framework to map applications over a packet-switched network of modules (single FPGA) and then to seamlessly partition such a network over multiple FPGAs over quasi-serial links. We illustrate the framework through three application case studies: LDPC Decoding, Particle Filter based Object Tracking, and Matrix Vector Multiplication over GF(2). Starting with high-level representations of each case application, we first express them in an intermediate message passing formulation, a model of communicating processing elements. Once the processing elements are identified, these are either handcrafted or realized using HLS. The rest of the flow is automated where the processing elements are plugged on to a configurable network-on-chip (CONNECT) topology of choice, followed by partitioning the 'on-chip' links to work seamlessly across chips/FPGAs.

cs.DC↗

arXiv2014

A Novel Approach to Develop a New Hybrid Technique for Trademark Image Retrieval

Trademark Image Retrieval is playing a vital role as a part of CBIR System. Trademark is of great significance because it carries the status value of any company. To retrieve such a fake or copied trademark we design a retrieval system which is based on hybrid techniques. It contains a mixture of two different feature vector which combined together to give a suitable retrieval system. In the proposed system we extract the corner feature which is applied on an edge pixel image. This feature is used to extract the relevant image and to more purify the result we apply other feature which is the invariant moment feature. From the experimental result we conclude that the system is 85 percent efficient.

cs.CV↗