Related papers: Optimizing Layer-Fused Scheduling of Transformer N…

Access to parallel and distributed computation has enabled researchers and developers to improve algorithms and performance in many applications. Recent research has focused on next generation special purpose systems with multiple kinds of…

As deep neural networks (DNNs) are being applied to a wide range of edge intelligent applications, it is critical for edge inference platforms to have both high-throughput and low-latency at the same time. Such edge platforms with multiple…

As machine learning techniques become ubiquitous, the efficiency of neural network implementations is becoming correspondingly paramount. Frameworks, such as Halide and TVM, separate out the algorithmic representation of the network from…

The need to execute Deep Neural Networks (DNNs) at low latency and low power at the edge has spurred the development of new heterogeneous Systems-on-Chips (SoCs) encapsulating a diverse set of hardware accelerators. How to optimally map a…

The scaling of computation throughput continues to outpace improvements in memory bandwidth, making many deep learning workloads memory-bound. Kernel fusion is a key technique to alleviate this problem, but the fusion strategies of existing…

We study the factors affecting training time in multi-device deep learning systems. Given a specification of a convolutional neural network, our goal is to minimize the time to train this model on a cluster of commodity CPUs and GPUs. We…

Edge devices like Nvidia Jetson platforms now offer several on-board accelerators -- including GPU CUDA cores, Tensor Cores, and Deep Learning Accelerators (DLA) -- which can be concurrently exploited to boost deep neural network (DNN)…

High-order tensor decomposition has been widely adopted to obtain compact deep neural networks for edge deployment. However, existing studies focus primarily on its algorithmic advantages such as accuracy and compression ratio-while…

Simulating physical systems using Partial Differential Equations (PDEs) has become an indispensible part of modern industrial process optimization. Traditionally, numerical solvers have been used to solve the associated PDEs, however…

The transformer has revolutionized modern AI across language, vision, and beyond. It consists of $L$ layers, each running $H$ attention heads in parallel and feeding the combined output to the subsequent layer. In attention, the input…

Transformer-based models have emerged as a leading architecture for natural language processing, natural language generation, and image generation tasks. A fundamental element of the transformer architecture is self-attention, which allows…

A key functionality of emerging connected autonomous systems such as smart transportation systems, smart cities, and the industrial Internet-of-Things, is the ability to process and learn from data collected at different physical locations.…

Spiking Neural Networks(SNNs) provide a brain-inspired and event-driven mechanism that is believed to be critical to unlock energy-efficient deep learning. The mixture-of-experts approach mirrors the parallel distributed processing of…

In-network computation represents a transformative approach to addressing the escalating demands of Artificial Intelligence (AI) workloads on network infrastructure. By leveraging the processing capabilities of network devices such as…

Transformer-based neural networks have achieved state-of-the-art task performance in a number of machine learning domains including natural language processing and computer vision. To further improve their accuracy, recent work has explored…

Modern Deep Neural Network (DNN) accelerators are equipped with increasingly larger on-chip buffers to provide more opportunities to alleviate the increasingly severe DRAM bandwidth pressure. However, most existing research on buffer…

The proliferation of IoT devices in smart cities challenges 6G networks with conflicting energy-latency requirements across heterogeneous slices. Existing approaches struggle with the energy-latency trade-off, particularly for massive scale…

We propose a novel GPU-cluster scheduler for distributed DL (DDL) workloads that enables proximity based consolidation of GPU resources based on the DDL jobs' sensitivities to the anticipated communication-network delays. Our scheduler…

Scheduling deep learning (DL) models to train on powerful clusters with accelerators like GPUs and TPUs, presently falls short, either lacking fine-grained heterogeneity awareness or leaving resources substantially under-utilized. To fill…

The use of multi-modal data for deep machine learning has shown promise when compared to uni-modal approaches with fusion of multi-modal features resulting in improved performance in several applications. However, most state-of-the-art…