An official website of the United States government
Spiking Neural Networks (SNNs) have gained popularity due to their high energy efficiency. Prior works have proposed various methods for training SNNs, including backpropagation-based methods. Training SNNs is computationally expensive compared to their conventional counterparts and would benefit from multiprocessor hardware acceleration. This is the first paper to propose inter-layer pipelining to accelerate training in SNNs using systolic array-based processors and multiprocessor scheduling. The impact of training using delayed gradients is observed using four networks training on different datasets, showing no degradation for small networks and < 10% degradation for large networks. The mapping of various training tasks of the SNN onto systolic arrays is formulated, and the proposed scheduling method is evaluated on the four networks. The results are compared against standard pipelining algorithms. The results show that the proposed method achieves an average speedup of 1.7× compared to standard pipelining algorithms, with an upwards of 2× improvement in some cases. The incurred communication overhead due to the proposed method is less than 0.5% of the total required communication of training in networks with convolutional layers.
more »
« less
LayerPipe: Accelerating Deep Neural Network Training by Intra-Layer and Inter-Layer Gradient Pipelining and Multiprocessor Scheduling
The time required for training the neural networks increases with size, complexity, and depth. Training model parameters by backpropagation inherently creates feedback loops. These loops hinder efficient pipelining and scheduling of the tasks within the layer and between consecutive layers. Prior approaches, such as PipeDream, have exploited the use of delayed gradient to achieve inter-layer pipelining. However, these approaches treat the entire backpropagation as a single task; this leads to an increase in computation time and processor underutilization. This paper presents novel optimization approaches where the gradient computations with respect to the weights and the activation functions are considered independently; therefore, these can be computed in parallel. This is referred to as intra-layer optimization. Additionally, the gradient computation with respect to the activation function is further divided into two parts and distributed to two consecutive layers. This leads to balanced scheduling where the computation time of each layer is the same. This is referred to as inter-layer optimization. The proposed system, referred to as LayerPipe, reduces the number of clock cycles required for training while maximizing processor utilization with minimal inter-processor communication overhead. LayerPipe achieves an average speedup of 25 % and upwards of 80% with 7 to 9 processors with less communication overhead when compared to PipeDream.
more »
« less
- Award ID(s):
- 1954749
- PAR ID:
- 10355269
- Date Published:
- Journal Name:
- 2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)
- Page Range / eLocation ID:
- 1 to 8
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
DNN training is extremely time-consuming, necessitating efficient multi-accelerator parallelization. Current approaches to parallelizing training primarily use intra-batch parallelization, where a single iteration of training is split over the available workers, but suffer from diminishing returns at higher worker counts. We present PipeDream, a system that adds inter-batch pipelining to intra-batch parallelism to further improve parallel training throughput, helping to better overlap computation with communication and reduce the amount of communication when possible. Unlike traditional pipelining, DNN training is bi-directional, where a forward pass through the computation graph is followed by a backward pass that uses state and intermediate data computed during the forward pass. Naïve pipelining can thus result in mismatches in state versions used in the forward and backward passes, or excessive pipeline flushes and lower hardware efficiency. To address these challenges, PipeDream versions model parameters for numerically correct gradient computations, and schedules forward and backward passes of different minibatches concurrently on different workers with minimal pipeline stalls. PipeDream also automatically partitions DNN layers among workers to balance work and minimize communication. Extensive experimentation with a range of DNN tasks, models, and hardware configurations shows that PipeDream trains models to high accuracy up to 5.3X faster than commonly used intra-batch parallelism techniques.more » « less
-
This paper addresses design of accelerators using systolic architectures for training of neural networks using a novel gradient interleaving approach. Training the neural network involves backpropagation of error and computation of gradients with respect to the activation functions and weights. It is shown that the gradient with respect to the activation function can be computed using a weight-stationary systolic array while the gradient with respect to the weights can be computed using an output-stationary systolic array. The novelty of the proposed approach lies in interleaving the computations of these two gradients to the same configurable systolic array. This results in reuse of the variables from one computation to the other and eliminates unnecessary memory accesses. The proposed approach leads to 1.4–2.2× savings in terms of number of cycles and 1.9× savings in terms of memory accesses. Thus, the proposed accelerator reduces latency and energy consumption.more » « less
-
This paper addresses the design of accelerators using systolic architectures to train convolutional neural networks using a novel gradient interleaving approach. Training the neural network involves computation and backpropagation of gradients of error with respect to the activation functions and weights. It is shown that the gradient with respect to the activation function can be computed using a weight-stationary systolic array, while the gradient with respect to the weights can be computed using an output-stationary systolic array. The novelty of the proposed approach lies in interleaving the computations of these two gradients on the same configurable systolic array. This results in the reuse of the variables from one computation to the other and eliminates unnecessary memory accesses and energy consumption associated with these memory accesses. The proposed approach leads to 1.4−2.2× savings in terms of the number of cycles and 1.9× savings in terms of memory accesses in the fully-connected layer. Furthermore, the proposed method uses up to 25% fewer cycles and memory accesses, and 16% less energy than baseline implementations for state-of-the-art CNNs. Under iso-area comparisons, for Inception-v4, compared to weight-stationary (WS), Intergrad achieves 12% savings in energy, 17% savings in memory, and 4% savings in cycles. Savings for Densenet-264 are 18% , 26% , and 27% with respect to energy, memory, and cycles, respectively. Thus, the proposed novel accelerator architecture reduces the latency and energy consumption for training deep neural networks.more » « less
-
The ever-increasing number of layers, millions of parameters, and large data volume make deep learning workloads resource-intensive and power-hungry. In this paper, we develop a convolutional neural network (CNN) acceleration framework, named MLCNN, which explores algorithm-hardware co-design to achieve cross-layer cooperative optimization and acceleration. MLCNN dramatically reduces computation and on-off chip communication, improving CNN’s performance. To achieve this, MLCNN reorders the position of nonlinear activation layers and pooling layers, which we prove results in a negligible accuracy loss; then the convolutional layer and pooling layer are cooptimized by means of redundant multiplication elimination, local addition reuse, and global addition reuse. To the best of our knowledge, MLCNN is the first of its kind that incorporates cooperative optimization across convolutional, activation, and pooling layers. We further customize the MLCNN accelerator to take full advantage of cross-layer CNN optimization to reduce both computation and on-off chip communication. Our analysis shows that MLCNN can significantly reduce (up to 98%) multiplications and additions. We have implemented a prototype of MLCNN and evaluated its performance on several widely used CNN models using both an accelerator-level cycle and energy model and RTL implementation. Experimental results show that MLCNN achieves 3.2× speedup and 2.9× energy efficiency compared with dense CNNs. MLCNN’s optimization methods are orthogonal to other CNN acceleration techniques, such as quantization and pruning. Combined with quantization, our quantized MLCNN gains a 12.8× speedup and 11.3× energy efficiency compared with DCNN.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICCAD51958.2021.9643567
