Domain-specific hardware is becoming a promising topic in the backdrop of improvement slow down for general-purpose processors due to the foreseeable end of Moore’s Law. Machine learning, especially deep neural networks (DNNs), has become the most dazzling domain witnessing successful applications in a wide spectrum of artificial intelligence (AI) tasks. The incomparable accuracy of DNNs is achieved by paying the cost of hungry memory consumption and high computational complexity, which greatly impedes their deployment in embedded systems. Therefore, the DNN compression concept was naturally proposed and widely used for memory saving and compute acceleration. In the past few years, a tremendous number of compression techniques have sprung up to pursue a satisfactory tradeoff between processing efficiency and application accuracy. Recently, this wave has spread to the design of neural network accelerators for gaining extremely high performance. However, the amount of related works is incredibly huge and the reported approaches are quite divergent. This research chaos motivates us to provide a comprehensive survey on the recent advances toward the goal of efficient compression and execution of DNNs without significantly compromising accuracy, involving both the high-level algorithms and their applications in hardware design. In this article, we review the mainstream compression approaches such as compact model, tensor decomposition, data quantization, and network sparsification. We explain their compression principles, evaluation metrics, sensitivity analysis, and joint-way use. Then, we answer the question of how to leverage these methods in the design of neural network accelerators and present the state-of-the-art hardware architectures. In the end, we discuss several existing issues such as fair comparison, testing workloads, automatic compression, influence on security, and framework/hardware-level support, and give promising topics in this field and the possible challenges as well. This article attempts to enable readers to quickly build up a big picture of neural network compression and acceleration, clearly evaluate various methods, and confidently get started in the right way.

Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey

Safety is the most important requirement for autonomous vehicles; hence, the ultimate challenge of designing an edge computing ecosystem for autonomous vehicles is to deliver enough computing power, redundancy, and security so as to guarantee the safety of autonomous vehicles. Specifically, autonomous driving systems are extremely complex; they tightly integrate many technologies, including sensing, localization, perception, decision making, as well as the smooth interactions with cloud platforms for high-definition (HD) map generation and data storage. These complexities impose numerous challenges for the design of autonomous driving edge computing systems. First, edge computing systems for autonomous driving need to process an enormous amount of data in real time, and often the incoming data from different sensors are highly heterogeneous. Since autonomous driving edge computing systems are mobile, they often have very strict energy consumption restrictions. Thus, it is imperative to deliver sufficient computing power with reasonable energy consumption, to guarantee the safety of autonomous vehicles, even at high speed. Second, in addition to the edge system design, vehicle-to-everything (V2X) provides redundancy for autonomous driving workloads and alleviates stringent performance and energy constraints on the edge side. With V2X, more research is required to define how vehicles cooperate with each other and the infrastructure. Last, safety cannot be guaranteed when security is compromised. Thus, protecting autonomous driving edge computing systems against attacks at different layers of the sensing and computing stack is of paramount concern. In this paper, we review state-of-the-art approaches in these areas as well as explore potential solutions to address these challenges.

Edge Computing for Autonomous Driving: Opportunities and Challenges

Deep neural networks (DNN) have demonstrated highly promising results across computer vision and speech recognition, and are becoming foundational for ubiquitous AI. The computational complexity of these algorithms and a need for high energy-efficiency has led to a surge in research on hardware accelerators. % for this paradigm. To reduce the latency and energy costs of accessing DRAM, most DNN accelerators are spatial in nature, with hundreds of processing elements (PE) operating in parallel and communicating with each other directly. DNNs are evolving at a rapid rate, and it is common to have convolution, recurrent, pooling, and fully-connected layers with varying input and filter sizes in the most recent topologies.They may be dense or sparse. They can also be partitioned in myriad ways (within and across layers) to exploit data reuse (weights and intermediate outputs). All of the above can lead to different dataflow patterns within the accelerator substrate. Unfortunately, most DNN accelerators support only fixed dataflow patterns internally as they perform a careful co-design of the PEs and the network-on-chip (NoC). In fact, the majority of them are only optimized for traffic within a convolutional layer. This makes it challenging to map arbitrary dataflows on the fabric efficiently, and can lead to underutilization of the available compute resources. DNN accelerators need to be programmable to enable mass deployment. For them to be programmable, they need to be configurable internally to support the various dataflow patterns that could be mapped over them. To address this need, we present MAERI, which is a DNN accelerator built with a set of modular and configurable building blocks that can easily support myriad DNN partitions and mappings by appropriately configuring tiny switches. MAERI provides 8-459% better utilization across multiple dataflow mappings over baselines with rigid NoC fabrics.

MAERI: Enabling Flexible Dataflow Mapping over DNN Accelerators via Reconfigurable Interconnects

Pruning and quantization for deep neural network acceleration: A survey

Brain-inspired hyperdimensional (HD) computing emulates cognition tasks by computing with hypervectors as an alternative to computing with numbers. At its very core, HD computing is about manipulating and comparing large patterns, stored in memory as hypervectors: the input symbols are mapped to a hypervector and an associative search is performed for reasoning and classification. For every classification event, an associative memory is in charge of finding the closest match between a set of learned hypervectors and a query hypervector by using a distance metric. Hypervectors with the i.i.d. components qualify a memory-centric architecture to tolerate massive number of errors, hence it eases cooperation of various methodological design approaches for boosting energy efficiency and scalability. This paper proposes architectural designs for hyperdimensional associative memory (HAM) to facilitate energy-efficient, fast, and scalable search operation using three widely-used design approaches. These HAM designs search for the nearest Hamming distance, and linearly scale with the number of dimensions in the hypervectors while exploring a large design space with orders of magnitude higher efficiency. First, we propose a digital CMOS-based HAM (D-HAM) that modularly scales to any dimension. Second, we propose a resistive HAM (R-HAM) that exploits timing discharge characteristic of nonvolatile resistive elements to approximately compute Hamming distances at a lower cost. Finally, we combine such resistive characteristic with a currentbased search method to design an analog HAM (A-HAM) that results in faster and denser alternative. Our experimental results show that R-HAM and A-HAM improve the energy-delay product by 9.6× and 1347× compared to D-HAM while maintaining a moderate accuracy of 94% in language recognition.

Exploring Hyperdimensional Associative Memory

Batch normalization (BN) has recently become a standard component for accelerating and improving the training of deep neural networks (DNNs). However, BN brings in additional calculations, consumes more memory, and significantly slows down the training iteration. Furthermore, the nonlinear square and sqrt operations in the normalization process impede low bit-width quantization techniques, which draw much attention to the deep learning hardware community. In this paper, we propose an $L1$ -norm BN (L1BN) with only linear operations in both forward and backward propagations during training. L1BN is approximately equivalent to the conventional $L2$ -norm BN (L2BN) by multiplying a scaling factor that equals $({\pi }/{2})^{1/2}$ . Experiments on various convolutional neural networks and generative adversarial networks reveal that L1BN can maintain the same performance and convergence rate as L2BN but with higher computational efficiency. In real application-specified integrated circuit synthesis with reduced resources, L1BN achieves 25% speedup and 37% energy saving compared to the original L2BN. Our hardware-friendly normalization method not only surpasses L2BN in speed but also simplifies the design of deep learning accelerators. Last but not least, L1BN promises a fully quantized training of DNNs, which empowers future artificial intelligence applications on mobile devices with transfer and continual learning capability.

https://par.nsf.gov/servlets/purl/10080632

$L1$ -Norm Batch Normalization for Efficient Training of Deep Neural Networks

3D die stacking and 2.5D interposer design are promising technologies to improve integration density, performance and cost. Current approaches face serious issues in dealing with emerging security challenges such as side channel attacks, hardware trojans, secure IC manufacturing and IP piracy. By utilizing intrinsic characteristics of 2.5D and 3D technologies, we propose novel opportunities in designing secure systems. We present: (i) a 3D architecture for shielding side-channel information; (ii) split fabrication using active interposers; (iii) circuit camouflage on monolithic 3D IC, and (iv) 3D IC-based security processing-in-memory (PIM). Advantages and challenges of these designs are discussed, showing that the new designs can improve existing counter-measures against security threats and further provide new security features.

Leveraging 3D Technologies for Hardware Security: Opportunities and Challenges

Ternary Content-Addressable Memory (TCAM) is widely used in networking routers, fully associative caches, search engines, etc. While the conventional SRAM-based TCAM suffers from the poor scalability, the emerging nonvolatile memories (NVM, i.e., MRAM, PCM, and ReRAM) bring evolution for the TCAM design. It effectively reduces the cell size, and makes significant energy reduction and scalability improvement. New applications such as associative processors/accelerators are facilitated by the emergence of the nonvolatile TCAM (nvTCAM). However, nvTCAM design is challenging. In addition to the emerging device's uncertainty, the nvTCAM cell structure is so diverse that it results in a design space too large to explore manually. To tackle these challenges, we propose a circuit-level model and develop a simulation tool, NVSim-CAM, which helps researchers to make early design decisions, and to evaluate device/circuit innovations. The tool is validated by HSPICE simulations and data from fabricated chips. We also present a case study to illustrate how NVSim-CAM benefits the nvTCAM design. In the case study, we propose a novel 3D vertical ReRAM based TCAM cell, the 3DvTCAM. We project the advantages/disadvantages and explore the design space for the proposed cell with NVSim-CAM.

https://dl.acm.org/doi/pdf/10.1145/2966986.2967059

NVSim-CAM: a circuit-level simulator for emerging nonvolatile memory based content-addressable memory

We propose to execute deep neural networks (DNNs) with dynamic and sparse graph (DSG) structure for compressive memory and accelerative execution during both training and inference. The great success of DNNs motivates the pursuing of lightweight models for the deployment onto embedded devices. However, most of the previous studies optimize for inference while neglect training or even complicate it. Training is far more intractable, since (i) the neurons dominate the memory cost rather than the weights in inference; (ii) the dynamic activation makes previous sparse acceleration via one-off optimization on fixed weight invalid; (iii) batch normalization (BN) is critical for maintaining accuracy while its activation reorganization damages the sparsity. To address these issues, DSG activates only a small amount of neurons with high selectivity at each iteration via a dimension-reduction search (DRS) and obtains the BN compatibility via a double-mask selection (DMS). Experiments show significant memory saving (1.7-4.5x) and operation reduction (2.3-4.4x) with little accuracy loss on various benchmarks.

/pdf/dynamic-sparse-graph-for-efficient-deep-learning-2hgpuyjzph.pdf

Dynamic Sparse Graph for Efficient Deep Learning

Spin-transfer torque magnetic random access memory (STT-RAM) technology has emerged as a potential replacement of SRAM in cache design, especially for building large-scale and energy-efficient last level caches. Compared with singlelevel cell (SLC), multi-level cell (MLC) STT-RAM is expected to double cache capacity and increase system performance. However, the two-step read/write access schemes incur considerable energy consumption and performance degradation. In this paper, we propose two techniques using data compression to optimize MLC STT-RAM cache design. The first technique tries to compress a cache line and fit it into only the soft-bit region of the cells, so that reading or writing this cache line takes only one step which is fast and energy-efficient. We introduce a second technique to increase the cache capacity by enabling the left hard-bit region to store another compressed cache line, which can improve the system performance for memory intensive workloads. The experimental results show that, compared with a conventional MLC STT-RAM last level cache design, our overhead minimized technique reduces the dynamic energy consumption by 38.2% on average with the same system performance, and our capacity augmented technique boosts the system performance by 6.1% with 19.2% dynamic energy saving on average, across the evaluated multi-programmed benchmarks.

Liu Liu

Papers

$L1$ -Norm Batch Normalization for Efficient Training of Deep Neural Networks

Leveraging 3D Technologies for Hardware Security: Opportunities and Challenges

NVSim-CAM: a circuit-level simulator for emerging nonvolatile memory based content-addressable memory

Dynamic Sparse Graph for Efficient Deep Learning

Building energy-efficient multi-level cell STT-RAM caches with data compression