UMass Amherst Engineers Use Memristors to Bypass Computing Bottleneck in Memory Critical for Artificial Intelligence
January 25, 2019 | UMass AmherstEstimated reading time: 2 minutes
Opening the way for advances in applications such as natural language understanding, machine translation, speech recognition and video surveillance, a team of researchers headed by Qiangfei Xia and Joshua Yang, electrical and computer engineers at the University of Massachusetts Amherst, says it can use memristor crossbar arrays to overcome a key bottleneck in traditional computing architecture.
The researchers say a memristor is a two-terminal “memory resistor” that performs computation at the same location where information is stored. This feature removes the need for traditional computers to transfer data between two sites, leading to a much better energy-speed efficiency. A memristor crossbar is a matrix of the tiny switches.
“Recent breakthroughs in recurrent deep neural networks with long short-term memory (LSTM) units have led to major advances in artificial intelligence,” the researchers say. However, state-of-the-art LSTM models with significantly increased complexity and a large number of parameters have a bottleneck in computing power resulting from both limited memory capacity and limited data communication capacity.
A solution to this LSTM blockage can be implemented with a memristor crossbar array, which has a small circuit footprint, can store a large number of parameters, and offers in-memory computing capability that contributes to circumventing what is known as the von Neumann bottleneck. “We illustrate the capability of our crossbar system as a core component in solving real-world problems, and show that memristor LSTM is a promising low-power and low-latency hardware platform for edge inference.”
The von Neumann bottleneck refers to the limits on the amount of data that can be transferred and energy efficiency in a computer built using the von Neumann architecture, in which the data processing and memory units are physically separated with a single common bus in between. John von Neumann was a 20th century mathematician, scientist, and computer science pioneer who in 1945 proposed the computer architecture which is still the basis for digital computers today.
“The memristor crossbar implementation of an LSTM,” say the authors, “to the best of our knowledge, has yet to be demonstrated, primarily because of the relative scarcity of large memristor arrays. In this work, we demonstrate our experimental implementation of a core part of LSTM networks in memristor crossbar arrays.”
As a demonstration, the authors applied the memristor-based LSTM in predicting the number of airline passengers based on data from past years, and in recognizing a person according to the way she or he walks. This is important in identifying a person when the face is camouflaged or facial recognition is technically difficult. “This work shows that the LSTM networks built in memristor crossbar arrays represent a promising alternative computing paradigm with high-speed energy efficiency.”
Suggested Items
Siemens Launches Digital Decarbonization Tool at Climate Week NYC to Help Simplify the Path to Net-Zero Facilities
09/27/2024 | BUSINESS WIRESiemens Financial Services launched the Decarbonization Business Optimizer (DBO), a cloud-based tool created to help address the overwhelming complexity of decarbonizing buildings and the financing needed to get to net-zero emissions.
Keysight to Facilitate UL Solutions Thunderbolt™ 5 Certification Testing
09/27/2024 | Keysight TechnologiesKeysight Technologies, Inc. (NYSE: KEYS) announces that UL Solutions has appointed it as the Thunderbolt 5 product certification test partner. This new role confirms UL Solution's status as an Intel-authorized certification lab for Thunderbolt 5 technologies.
Flex Named to TIME's World's Best Companies List for Second Consecutive Year
09/26/2024 | FlexFlex announced its inclusion on TIME's 2024 World's Best Companies list, a comprehensive analysis conducted to identify the top performing companies across the globe.
Jabil Expands Silicon Photonics Capabilities to Enable Next-Generation AI and Data Center Technologies
09/26/2024 | JabilJabil Inc. a global leader in design, manufacturing, and supply chain solutions, announced its continued investment in silicon photonics-based products and capabilities to support the increasing demands of hyperscalers and next-wave cloud and AI data center growth.
Tech Trajectories: Projecting Trend Advancements
09/25/2024 | Srikanth Chandrasekaran, IEEE SAReflecting on technological advancements and examining how foundational technology trends evolve is essential to seeing where the world and society are headed. Last year, significant strides were made in data governance, automation, edge computing, artificial intelligence (AI), and the metaverse. IEEE Standards Association (IEEE SA) is keeping its eye on what’s ahead for these technologies that are transforming various industry sectors. In this article, I’ll examine the current state of these technologies and outline the advancements and directions we expect ahead.