Zobrazeno 1 - 10
of 197
pro vyhledávání: '"Emer, Joel"'
Latency and energy consumption are key metrics in the performance of deep neural network (DNN) accelerators. A significant factor contributing to latency and energy is data transfers. One method to reduce transfers or data is reusing data when multip
Externí odkaz:
http://arxiv.org/abs/2409.13625
This paper introduces the continuous tensor abstraction, allowing indices to take real-number values (e.g., A[3.14]), and provides a continuous loop construct that iterates over the infinitely large set of real numbers. This paper expands the existin
Externí odkaz:
http://arxiv.org/abs/2407.01742
Autor:
Nayak, Nandeeka, Wu, Xinrui, Odemuyiwa, Toluwanimi O., Pellauer, Michael, Emer, Joel S., Fletcher, Christopher W.
Attention for transformers is a critical workload that has recently received significant "attention" as a target for custom acceleration. Yet, while prior work succeeds in reducing attention's memory-bandwidth requirements, it creates load imbalance
Externí odkaz:
http://arxiv.org/abs/2406.10491
Autor:
Andrulis, Tanner, Chaudhry, Gohar Irfan, Suriyakumar, Vinith M., Emer, Joel S., Sze, Vivienne
Photonics is a promising technology to accelerate Deep Neural Networks as it can use optical interconnects to reduce data movement energy and it enables low-energy, high-throughput optical-analog computations. To realize these benefits in a full syst
Externí odkaz:
http://arxiv.org/abs/2405.07266
Compute-In-Memory (CiM) is a promising solution to accelerate Deep Neural Networks (DNNs) as it can avoid energy-intensive DNN weight movement and use memory arrays to perform low-energy, high-density computations. These benefits have inspired resear
Externí odkaz:
http://arxiv.org/abs/2405.07259
In this work, we propose a unified abstraction for graph algorithms: the Extended General Einsums language, or EDGE. The EDGE language expresses graph algorithms in the language of tensor algebra, providing a rigorous, succinct, and expressive mathem
Externí odkaz:
http://arxiv.org/abs/2404.11591
Analog Compute-in-Memory (CiM) accelerators use analog-digital converters (ADCs) to read the analog values that they compute. ADCs can consume significant energy and area, so architecture-level ADC decisions such as ADC resolution or number of ADCs c
Externí odkaz:
http://arxiv.org/abs/2404.06553
Publikováno v:
56th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO '23), 2023
Sparse tensor algebra is a challenging class of workloads to accelerate due to low arithmetic intensity and varying sparsity patterns. Prior sparse tensor algebra accelerators have explored tiling sparse data to increase exploitable data reuse and im
Externí odkaz:
http://arxiv.org/abs/2310.00192
This paper provides the first systematic analysis of a synergistic threat model encompassing memory corruption vulnerabilities and microarchitectural side-channel vulnerabilities. We study speculative shield bypass attacks that leverage speculative e
Externí odkaz:
http://arxiv.org/abs/2309.04119
Autor:
Wu, Yannan Nellie, Tsai, Po-An, Muralidharan, Saurav, Parashar, Angshuman, Sze, Vivienne, Emer, Joel S.
Due to complex interactions among various deep neural network (DNN) optimization techniques, modern DNNs can have weights and activations that are dense or sparse with diverse sparsity degrees. To offer a good trade-off between accuracy and hardware
Externí odkaz:
http://arxiv.org/abs/2305.12718