arXiv subjects

Thomas Parnell

Publications and source records attributed to Thomas Parnell.

31 records · Page 2Linked to original sources

arXiv2019

5 Parallel Prism: A topology for pipelined implementations of convolutional neural networks using computational memory

In-memory computing is an emerging computing paradigm that could enable deeplearning inference at significantly higher energy efficiency and reduced latency. The essential idea is to map the synaptic weights corresponding to each layer to one or more computational memory (CM) cores. During inference, these cores perform the associated matrix-vector multiply operations in place with O(1) time complexity, thus obviating the need to move the synaptic weights to an additional processing unit. Moreover, this architecture could enable the execution of these networks in a highly pipelined fashion. However, a key challenge is to design an efficient communication fabric for the CM cores. Here, we present one such communication fabric based on a graph topology that is well suited for the widely successful convolutional neural networks (CNNs). We show that this communication fabric facilitates the pipelined execution of all state of-the-art CNNs by proving the existence of a homomorphism between one graph representation of these networks and the proposed graph topology. We then present a quantitative comparison with established communication topologies and show that our proposed topology achieves the lowest bandwidth requirements per communication channel. Finally, we present a concrete example of mapping ResNet-32 onto an array of CM cores.

cs.LG↗

arXiv2019

Neuromorphic computing with multi-memristive synapses

Neuromorphic computing has emerged as a promising avenue towards building the next generation of intelligent computing systems. It has been proposed that memristive devices, which exhibit history-dependent conductivity modulation, could efficiently represent the synaptic weights in artificial neural networks. However, precise modulation of the device conductance over a wide dynamic range, necessary to maintain high network accuracy, is proving to be challenging. To address this, we present a multi-memristive synaptic architecture with an efficient global counter-based arbitration scheme. We focus on phase change memory devices, develop a comprehensive model and demonstrate via simulations the effectiveness of the concept for both spiking and non-spiking neural networks. Moreover, we present experimental results involving over a million phase change memory devices for unsupervised learning of temporal correlations using a spiking neural network. The work presents a significant step towards the realization of large-scale and energy-efficient neuromorphic computing systems.

cs.ET↗

arXiv2019

Benchmarking and Optimization of Gradient Boosting Decision Tree Algorithms

Gradient boosting decision trees (GBDTs) have seen widespread adoption in academia, industry and competitive data science due to their state-of-the-art performance in many machine learning tasks. One relative downside to these models is the large number of hyper-parameters that they expose to the end-user. To maximize the predictive power of GBDT models, one must either manually tune the hyper-parameters, or utilize automated techniques such as those based on Bayesian optimization. Both of these approaches are time-consuming since they involve repeatably training the model for different sets of hyper-parameters. A number of software GBDT packages have started to offer GPU acceleration which can help to alleviate this problem. In this paper, we consider three such packages: XGBoost, LightGBM and Catboost. Firstly, we evaluate the performance of the GPU acceleration provided by these packages using large-scale datasets with varying shapes, sparsities and learning tasks. Then, we compare the packages in the context of hyper-parameter optimization, both in terms of how quickly each package converges to a good validation score, and in terms of generalization performance.

cs.LG↗

arXiv2018

Parallel training of linear models without compromising convergence

In this paper we analyze, evaluate, and improve the performance of training generalized linear models on modern CPUs. We start with a state-of-the-art asynchronous parallel training algorithm, identify system-level performance bottlenecks, and apply optimizations that improve data parallelism, cache line locality, and cache line prefetching of the algorithm. These modifications reduce the per-epoch run-time significantly, but take a toll on algorithm convergence in terms of the required number of epochs. To alleviate these shortcomings of our systems-optimized version, we propose a novel, dynamic data partitioning scheme across threads which allows us to approach the convergence of the sequential version. The combined set of optimizations result in a consistent bottom line speedup in convergence of up to 12x compared to the initial asynchronous parallel training algorithm and up to 42x, compared to state of the art implementations (scikit-learn and h2o) on a range of multi-core CPU architectures.

cs.LG↗

arXiv2018

Snap ML: A Hierarchical Framework for Machine Learning

We describe a new software framework for fast training of generalized linear models. The framework, named Snap Machine Learning (Snap ML), combines recent advances in machine learning systems and algorithms in a nested manner to reflect the hierarchical architecture of modern computing systems. We prove theoretically that such a hierarchical system can accelerate training in distributed environments where intra-node communication is cheaper than inter-node communication. Additionally, we provide a review of the implementation of Snap ML in terms of GPU acceleration, pipelining, communication patterns and software architecture, highlighting aspects that were critical for achieving high performance. We evaluate the performance of Snap ML in both single-node and multi-node environments, quantifying the benefit of the hierarchical scheme and the data streaming functionality, and comparing with other widely-used machine learning software frameworks. Finally, we present a logistic regression benchmark on the Criteo Terabyte Click Logs dataset and show that Snap ML achieves the same test loss an order of magnitude faster than any of the previously reported results, including those obtained using TensorFlow and scikit-learn.

cs.LG↗

arXiv2018

Elastic CoCoA: Scaling In to Improve Convergence

In this paper we experimentally analyze the convergence behavior of CoCoA and show, that the number of workers required to achieve the highest convergence rate at any point in time, changes over the course of the training. Based on this observation, we build Chicle, an elastic framework that dynamically adjusts the number of workers based on feedback from the training algorithm, in order to select the number of workers that results in the highest convergence rate. In our evaluation of 6 datasets, we show that Chicle is able to accelerate the time-to-accuracy by a factor of up to 5.96x compared to the best static setting, while being robust enough to find an optimal or near-optimal setting automatically in most cases.

cs.LG↗

arXiv2017

Understanding and Optimizing the Performance of Distributed Machine Learning Applications on Apache Spark

In this paper we explore the performance limits of Apache Spark for machine learning applications. We begin by analyzing the characteristics of a state-of-the-art distributed machine learning algorithm implemented in Spark and compare it to an equivalent reference implementation using the high performance computing framework MPI. We identify critical bottlenecks of the Spark framework and carefully study their implications on the performance of the algorithm. In order to improve Spark performance we then propose a number of practical techniques to alleviate some of its overheads. However, optimizing computational efficiency and framework related overheads is not the only key to performance -- we demonstrate that in order to get the best performance out of any implementation it is necessary to carefully tune the algorithm to the respective trade-off between computation time and communication latency. The optimal trade-off depends on both the properties of the distributed algorithm as well as infrastructure and framework-related characteristics. Finally, we apply these technical and algorithmic optimizations to three different distributed linear machine learning algorithms that have been implemented in Spark. We present results using five large datasets and demonstrate that by using the proposed optimizations, we can achieve a reduction in the performance difference between Spark and MPI from 20x to 2x.

cs.DC↗

arXiv2017

Linear-Complexity Relaxed Word Mover's Distance with GPU Acceleration

The amount of unstructured text-based data is growing every day. Querying, clustering, and classifying this big data requires similarity computations across large sets of documents. Whereas low-complexity similarity metrics are available, attention has been shifting towards more complex methods that achieve a higher accuracy. In particular, the Word Mover's Distance (WMD) method proposed by Kusner et al. is a promising new approach, but its time complexity grows cubically with the number of unique words in the documents. The Relaxed Word Mover's Distance (RWMD) method, again proposed by Kusner et al., reduces the time complexity from qubic to quadratic and results in a limited loss in accuracy compared with WMD. Our work contributes a low-complexity implementation of the RWMD that reduces the average time complexity to linear when operating on large sets of documents. Our linear-complexity RWMD implementation, henceforth referred to as LC-RWMD, maps well onto GPUs and can be efficiently distributed across a cluster of GPUs. Our experiments on real-life datasets demonstrate 1) a performance improvement of two orders of magnitude with respect to our GPU-based distributed implementation of the quadratic RWMD, and 2) a performance improvement of three to four orders of magnitude with respect to our distributed WMD implementation that uses GPU-based RWMD for pruning.

cs.IR↗

arXiv2017

Efficient Use of Limited-Memory Accelerators for Linear Learning on Heterogeneous Systems

We propose a generic algorithmic building block to accelerate training of machine learning models on heterogeneous compute systems. Our scheme allows to efficiently employ compute accelerators such as GPUs and FPGAs for the training of large-scale machine learning models, when the training data exceeds their memory capacity. Also, it provides adaptivity to any system's memory hierarchy in terms of size and processing speed. Our technique is built upon novel theoretical insights regarding primal-dual coordinate methods, and uses duality gap information to dynamically decide which part of the data should be made available for fast processing. To illustrate the power of our approach we demonstrate its performance for training of generalized linear models on a large-scale dataset exceeding the memory size of a modern GPU, showing an order-of-magnitude speedup over existing approaches.

cs.LG↗

arXiv2017

Temporal correlation detection using computational phase-change memory

For decades, conventional computers based on the von Neumann architecture have performed computation by repeatedly transferring data between their processing and their memory units, which are physically separated. As computation becomes increasingly data-centric and as the scalability limits in terms of performance and power are being reached, alternative computing paradigms are searched for in which computation and storage are collocated. A fascinating new approach is that of computational memory where the physics of nanoscale memory devices are used to perform certain computational tasks within the memory unit in a non-von Neumann manner. Here we present a large-scale experimental demonstration using one million phase-change memory devices organized to perform a high-level computational primitive by exploiting the crystallization dynamics. Also presented is an application of such a computational memory to process real-world data-sets. The results show that this co-existence of computation and storage at the nanometer scale could be the enabler for new, ultra-dense, low power, and massively parallel computing systems.

cs.ET↗

arXiv2017

Scalable and interpretable product recommendations via overlapping co-clustering

We consider the problem of generating interpretable recommendations by identifying overlapping co-clusters of clients and products, based only on positive or implicit feedback. Our approach is applicable on very large datasets because it exhibits almost linear complexity in the input examples and the number of co-clusters. We show, both on real industrial data and on publicly available datasets, that the recommendation accuracy of our algorithm is competitive to that of state-of-art matrix factorization techniques. In addition, our technique has the advantage of offering recommendations that are textually and visually interpretable. Finally, we examine how to implement our technique efficiently on Graphical Processing Units (GPUs).

cs.IR↗

arXiv2017

Large-Scale Stochastic Learning using GPUs

In this work we propose an accelerated stochastic learning system for very large-scale applications. Acceleration is achieved by mapping the training algorithm onto massively parallel processors: we demonstrate a parallel, asynchronous GPU implementation of the widely used stochastic coordinate descent/ascent algorithm that can provide up to 35x speed-up over a sequential CPU implementation. In order to train on very large datasets that do not fit inside the memory of a single GPU, we then consider techniques for distributed stochastic learning. We propose a novel method for optimally aggregating model updates from worker nodes when the training data is distributed either by example or by feature. Using this technique, we demonstrate that one can scale out stochastic learning across up to 8 worker nodes without any significant loss of training time. Finally, we combine GPU acceleration with the optimized distributed method to train on a dataset consisting of 200 million training examples and 75 million features. We show by scaling out across 4 GPUs, one can attain a high degree of training accuracy in around 4 seconds: a 20x speed-up in training time compared to a multi-threaded, distributed implementation across 4 CPUs.

cs.LG↗

arXiv2011

Information theory of massively parallel probe storage channels

Motivated by the concept of probe storage, we study the problem of information retrieval using a large array of N nano-mechanical probes, N ~ 4000. At the nanometer scale it is impossible to avoid errors in the positioning of the array, thus all signals retrieved by the probes of the array at a given sampling moment are affected by the same amount of random position jitter. Therefore a massively parallel probe storage device is an example of a noisy communication channel with long range correlations between channel outputs due to the global positioning errors. We find that these correlations have a profound effect on the channel's properties. For example, it turns out that the channel's information capacity does approach 1 bit per probe in the limit of high signal-to-noise ratio, but the rate of the approach is only polynomial in the channel noise strength. Moreover, any error correction code with block size N >> 1 such that codewords correspond to the instantaneous outputs of the all probes in the array exhibits an error floor independently of the code rate. We illustrate this phenomenon explicitly using Reed-Solomon codes the performance of which is easy to simulate numerically. We also discuss capacity-achieving error correction codes for the global jitter channel and their complexity.

cs.IT↗