arXiv subjects

Minsik Cho

Publications and source records attributed to Minsik Cho.

49 records · Page 3Linked to original sources

arXiv2022

DKM: Differentiable K-Means Clustering Layer for Neural Network Compression

Deep neural network (DNN) model compression for efficient on-device inference is becoming increasingly important to reduce memory requirements and keep user data on-device. To this end, we propose a novel differentiable k-means clustering layer (DKM) and its application to train-time weight clustering-based DNN model compression. DKM casts k-means clustering as an attention problem and enables joint optimization of the DNN parameters and clustering centroids. Unlike prior works that rely on additional regularizers and parameters, DKM-based compression keeps the original loss function and model architecture fixed. We evaluated DKM-based compression on various DNN models for computer vision and natural language processing (NLP) tasks. Our results demonstrate that DKM delivers superior compression and accuracy trade-off on ImageNet1k and GLUE benchmarks. For example, DKM-based compression can offer 74.5% top-1 ImageNet1k accuracy on ResNet50 DNN model with 3.3MB model size (29.4x model compression factor). For MobileNet-v1, which is a challenging DNN to compress, DKM delivers 63.9% top-1 ImageNet1k accuracy with 0.72 MB model size (22.4x model compression factor). This result is 6.8% higher top-1accuracy and 33% relatively smaller model size than the current state-of-the-art DNN compression algorithms. Additionally, DKM enables compression of DistilBERT model by 11.8x with minimal (1.1%) accuracy loss on GLUE NLP benchmarks.

cs.LG↗

arXiv2021

Exploring Avenues Beyond Revised DSD Functionals: I. range separation, with xDSD as a special case

We have explored the use of range separation as a possible avenue for further improvement on our revDSD minimally empirical double hybrid functionals. Such $ω$DSD functionals encompass the XYG3 type of double hybrid (i.e., xDSD) as a special case for $ω$->0. As in our previous studies, the large and chemically diverse GMTKN55 benchmark suite was used for evaluation. Especially when using the D4 rather than D3BJ dispersion model, xDSD has a slight performance advantage in WTMAD2. As found previously, PBEP86 is the winning combination for the semilocal parts. xDSDn-PBEP86-D4 marginally outperforms the previous 'best in class' $ω$B97M(2) Berkeley double hybrid, but without range separation and using fewer than half the number of empirical parameters. Range separation turns out to offer only marginal further improvements on GMTKN55 itself. While $ω$B97M(2) still yields better performance for small-molecule thermochemistry, this is outweighed in WTMAD2 by superior performance of the new functionals for conformer equilibria. Results for two external test sets with pronounced static correlation effects may indicate that range-separated double hybrids are more resilient to such effects.

physics.chem-ph↗

arXiv2021

NASTransfer: Analyzing Architecture Transferability in Large Scale Neural Architecture Search

Neural Architecture Search (NAS) is an open and challenging problem in machine learning. While NAS offers great promise, the prohibitive computational demand of most of the existing NAS methods makes it difficult to directly search the architectures on large-scale tasks. The typical way of conducting large scale NAS is to search for an architectural building block on a small dataset (either using a proxy set from the large dataset or a completely different small scale dataset) and then transfer the block to a larger dataset. Despite a number of recent results that show the promise of transfer from proxy datasets, a comprehensive evaluation of different NAS methods studying the impact of different source datasets has not yet been addressed. In this work, we propose to analyze the architecture transferability of different NAS methods by performing a series of experiments on large scale benchmarks such as ImageNet1K and ImageNet22K. We find that: (i) The size and domain of the proxy set does not seem to influence architecture performance on the target dataset. On average, transfer performance of architectures searched using completely different small datasets (e.g., CIFAR10) perform similarly to the architectures searched directly on proxy target datasets. However, design of proxy sets has considerable impact on rankings of different NAS methods. (ii) While different NAS methods show similar performance on a source dataset (e.g., CIFAR10), they significantly differ on the transfer performance to a large dataset (e.g., ImageNet1K). (iii) Even on large datasets, random sampling baseline is very competitive, but the choice of the appropriate combination of proxy set and search strategy can provide significant improvement over it. We believe that our extensive empirical analysis will prove useful for future design of NAS algorithms.

cs.CV↗

arXiv2020

Large Scale Neural Architecture Search with Polyharmonic Splines

Neural Architecture Search (NAS) is a powerful tool to automatically design deep neural networks for many tasks, including image classification. Due to the significant computational burden of the search phase, most NAS methods have focused so far on small, balanced datasets. All attempts at conducting NAS at large scale have employed small proxy sets, and then transferred the learned architectures to larger datasets by replicating or stacking the searched cells. We propose a NAS method based on polyharmonic splines that can perform search directly on large scale, imbalanced target datasets. We demonstrate the effectiveness of our method on the ImageNet22K benchmark[16], which contains 14 million images distributed in a highly imbalanced manner over 21,841 categories. By exploring the search space of the ResNet [23] and Big-Little Net ResNext [11] architectures directly on ImageNet22K, our polyharmonic splines NAS method designed a model which achieved a top-1 accuracy of 40.03% on ImageNet22K, an absolute improvement of 3.13% over the state of the art with similar global batch size [15].

cs.CV↗

arXiv2020

SimEx: Express Prediction of Inter-dataset Similarity by a Fleet of Autoencoders

Knowing the similarity between sets of data has a number of positive implications in training an effective model, such as assisting an informed selection out of known datasets favorable to model transfer or data augmentation problems with an unknown dataset. Common practices to estimate the similarity between data include comparing in the original sample space, comparing in the embedding space from a model performing a certain task, or fine-tuning a pretrained model with different datasets and evaluating the performance changes therefrom. However, these practices would suffer from shallow comparisons, task-specific biases, or extensive time and computations required to perform comparisons. We present SimEx, a new method for early prediction of inter-dataset similarity using a set of pretrained autoencoders each of which is dedicated to reconstructing a specific part of known data. Specifically, our method takes unknown data samples as input to those pretrained autoencoders, and evaluate the difference between the reconstructed output samples against their original input samples. Our intuition is that, the more similarity exists between the unknown data samples and the part of known data that an autoencoder was trained with, the better chances there could be that this autoencoder makes use of its trained knowledge, reconstructing output samples closer to the originals. We demonstrate that our method achieves more than 10x speed-up in predicting inter-dataset similarity compared to common similarity-estimating practices. We also demonstrate that the inter-dataset similarity estimated by our method is well-correlated with common practices and outperforms the baselines approaches of comparing at sample- or embedding-spaces, without newly training anything at the comparison time.

cs.LG↗

arXiv2019

Enabling real-time multi-messenger astrophysics discoveries with deep learning

Multi-messenger astrophysics is a fast-growing, interdisciplinary field that combines data, which vary in volume and speed of data processing, from many different instruments that probe the Universe using different cosmic messengers: electromagnetic waves, cosmic rays, gravitational waves and neutrinos. In this Expert Recommendation, we review the key challenges of real-time observations of gravitational wave sources and their electromagnetic and astroparticle counterparts, and make a number of recommendations to maximize their potential for scientific discovery. These recommendations refer to the design of scalable and computationally efficient machine learning algorithms; the cyber-infrastructure to numerically simulate astrophysical sources, and to process and interpret multi-messenger astrophysics data; the management of gravitational wave detections to trigger real-time alerts for electromagnetic and astroparticle follow-ups; a vision to harness future developments of machine learning and cyber-infrastructure resources to cope with the big-data requirements; and the need to build a community of experts to realize the goals of multi-messenger astrophysics.

gr-qc↗

arXiv2019

MUTE: Data-Similarity Driven Multi-hot Target Encoding for Neural Network Design

Target encoding is an effective technique to deliver better performance for conventional machine learning methods, and recently, for deep neural networks as well. However, the existing target encoding approaches require significant increase in the learning capacity, thus demand higher computation power and more training data. In this paper, we present a novel and efficient target encoding scheme, MUTE to improve both generalizability and robustness of a target model by understanding the inter-class characteristics of a target dataset. By extracting the confusion level between the target classes in a dataset, MUTE strategically optimizes the Hamming distances among target encoding. Such optimized target encoding offers higher classification strength for neural network models with negligible computation overhead and without increasing the model size. When MUTE is applied to the popular image classification networks and datasets, our experimental results show that MUTE offers better generalization and defense against the noises and adversarial attacks over the existing solutions.

cs.LG↗

arXiv2019

A Unified Approximation Framework for Compressing and Accelerating Deep Neural Networks

Deep neural networks (DNNs) have achieved significant success in a variety of real world applications, i.e., image classification. However, tons of parameters in the networks restrict the efficiency of neural networks due to the large model size and the intensive computation. To address this issue, various approximation techniques have been investigated, which seek for a light weighted network with little performance degradation in exchange of smaller model size or faster inference. Both low-rankness and sparsity are appealing properties for the network approximation. In this paper we propose a unified framework to compress the convolutional neural networks (CNNs) by combining these two properties, while taking the nonlinear activation into consideration. Each layer in the network is approximated by the sum of a structured sparse component and a low-rank component, which is formulated as an optimization problem. Then, an extended version of alternating direction method of multipliers (ADMM) with guaranteed convergence is presented to solve the relaxed optimization problem. Experiments are carried out on VGG-16, AlexNet and GoogLeNet with large image classification datasets. The results outperform previous work in terms of accuracy degradation, compression rate and speedup ratio. The proposed method is able to remarkably compress the model (with up to 4.9x reduction of parameters) at a cost of little loss or without loss on accuracy.

cs.LG↗

arXiv2019

Deep Learning for Multi-Messenger Astrophysics: A Gateway for Discovery in the Big Data Era

This report provides an overview of recent work that harnesses the Big Data Revolution and Large Scale Computing to address grand computational challenges in Multi-Messenger Astrophysics, with a particular emphasis on real-time discovery campaigns. Acknowledging the transdisciplinary nature of Multi-Messenger Astrophysics, this document has been prepared by members of the physics, astronomy, computer science, data science, software and cyberinfrastructure communities who attended the NSF-, DOE- and NVIDIA-funded "Deep Learning for Multi-Messenger Astrophysics: Real-time Discovery at Scale" workshop, hosted at the National Center for Supercomputing Applications, October 17-19, 2018. Highlights of this report include unanimous agreement that it is critical to accelerate the development and deployment of novel, signal-processing algorithms that use the synergy between artificial intelligence (AI) and high performance computing to maximize the potential for scientific discovery with Multi-Messenger Astrophysics. We discuss key aspects to realize this endeavor, namely (i) the design and exploitation of scalable and computationally efficient AI algorithms for Multi-Messenger Astrophysics; (ii) cyberinfrastructure requirements to numerically simulate astrophysical sources, and to process and interpret Multi-Messenger Astrophysics data; (iii) management of gravitational wave detections and triggers to enable electromagnetic and astro-particle follow-ups; (iv) a vision to harness future developments of machine and deep learning and cyberinfrastructure resources to cope with the scale of discovery in the Big Data Era; (v) and the need to build a community that brings domain experts together with data scientists on equal footing to maximize and accelerate discovery in the nascent field of Multi-Messenger Astrophysics.

astro-ph.IM↗

arXiv2018

Data-parallel distributed training of very large models beyond GPU capacity

GPUs have limited memory and it is difficult to train wide and/or deep models that cause the training process to go out of memory. It is shown in this paper how an open source tool called Large Model Support (LMS) can utilize a high bandwidth NVLink connection between CPUs and GPUs to accomplish training of deep convolutional networks. LMS performs tensor swapping between CPU memory and GPU memory such that only a minimal number of tensors required in a training step are kept in the GPU memory. It is also shown how LMS can be combined with an MPI based distributed deep learning module to train models in a data-parallel fashion across multiple GPUs, such that each GPU is utilizing the CPU memory for tensor swapping. The hardware architecture that enables the high bandwidth GPU link with the CPU is discussed as well as the associated set of software tools that are available as the PowerAI package.

cs.DC↗

arXiv2018

Ab Initio Finite Temperature Auxiliary Field Quantum Monte Carlo

We present an \textit{ab initio} auxiliary field quantum Monte Carlo method for studying the electronic structure of molecules, solids, and model Hamiltonians at finite temperature. The algorithm marries the \textit{ab initio} phaseless auxiliary field quantum Monte Carlo algorithm known to produce high accuracy ground state energies of molecules and solids with its finite temperature variant, long used by condensed matter physicists for studying model Hamiltonian phase diagrams, to yield a phaseless, \textit{ab initio} finite temperature method. We demonstrate that the method produces internal energies within chemical accuracy of exact diagonalization results across a wide range of temperatures for H$_{2}$O (STO-3G), C$_{2}$ (STO-6G), the one-dimensional hydrogen chain (STO-6G), and the multi-orbital Hubbard model. Our method effectively controls the phase problem through importance sampling, \textit{often even without invoking the phaseless approximation}, down to temperatures at which the systems studied approach their ground states and may therefore be viewed as exact over wide temperature ranges. This technique embodies a versatile tool for studying the finite temperature phase diagrams of a plethora of systems whose properties cannot be captured by a Hubbard U term alone. Our results moreover illustrate that the severity of the phase problem for model Hamiltonians far exceeds that for many molecules at all of the temperatures studied.

cond-mat.str-el↗

arXiv2017

PowerAI DDL

As deep neural networks become more complex and input datasets grow larger, it can take days or even weeks to train a deep neural network to the desired accuracy. Therefore, distributed Deep Learning at a massive scale is a critical capability, since it offers the potential to reduce the training time from weeks to hours. In this paper, we present a software-hardware co-optimized distributed Deep Learning system that can achieve near-linear scaling up to hundreds of GPUs. The core algorithm is a multi-ring communication pattern that provides a good tradeoff between latency and bandwidth and adapts to a variety of system configurations. The communication algorithm is implemented as a library for easy use. This library has been integrated into Tensorflow, Caffe, and Torch. We train Resnet-101 on Imagenet 22K with 64 IBM Power8 S822LC servers (256 GPUs) in about 7 hours to an accuracy of 33.8 % validation accuracy. Microsoft's ADAM and Google's DistBelief results did not reach 30 % validation accuracy for Imagenet 22K. Compared to Facebook AI Research's recent paper on 256 GPU training, we use a different communication algorithm, and our combined software and hardware system offers better communication overhead for Resnet-50. A PowerAI DDL enabled version of Torch completed 90 epochs of training on Resnet 50 for 1K classes in 50 minutes using 64 IBM Power8 S822LC servers (256 GPUs).

cs.DC↗

arXiv2017

MEC: Memory-efficient Convolution for Deep Neural Network

Convolution is a critical component in modern deep neural networks, thus several algorithms for convolution have been developed. Direct convolution is simple but suffers from poor performance. As an alternative, multiple indirect methods have been proposed including im2col-based convolution, FFT-based convolution, or Winograd-based algorithm. However, all these indirect methods have high memory-overhead, which creates performance degradation and offers a poor trade-off between performance and memory consumption. In this work, we propose a memory-efficient convolution or MEC with compact lowering, which reduces memory-overhead substantially and accelerates convolution process. MEC lowers the input matrix in a simple yet efficient/compact way (i.e., much less memory-overhead), and then executes multiple small matrix multiplications in parallel to get convolution completed. Additionally, the reduced memory footprint improves memory sub-system efficiency, improving performance. Our experimental results show that MEC reduces memory consumption significantly with good speedup on both mobile and server platforms, compared with other indirect convolution algorithms.

cs.LG↗