检索结果-内蒙古大学图书馆

Memory Organization for Energy-Efficient Learning and Inference in Digital Neuromorphic Accelerators

学校读者我要写书评

暂无评论

Memory Organization for Energy-Efficient Learning and Infere...

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Clemens J. S. Schaefer Patrick Faley Emre O. Neftci Siddharth Joshi Department of Computer Science and Engineering University of Notre Dame Notre Dame IN USA Department of Cognitive Sciences and Department of Computer Science UC Irvine Irvine CA USA

ISBN: (数字)9781728133201

ISBN: (纸本)9781728133218

The energy efficiency of neuromorphic hardware is greatly affected by the energy of storing, accessing, and updating synaptic parameters. Various methods of memory organisation targeting energy-efficient digital accelerators have been investigated in the past, however, they do not completely encapsulate the energy costs at a system level. To address this shortcoming and to account for various overheads, we synthesize the controller and memory for different encoding schemes and extract the energy costs from these synthesized blocks. Additionally, we introduce functional encoding for structured connectivity such as the connectivity in convolutional layers. Functional encoding offers a 58% reduction in the energy to implement a backward pass and weight update in such layers compared to existing index-based solutions. We show that for a 2 layer spiking neural network trained to retain a spatio-temporal pattern, bitmap (PB-BMP) based organization can encode the sparser networks more efficiently. This form of encoding delivers a 1.37× improvement in energy efficiency coming at the cost of a 4% degradation in network retention accuracy as measured by the van Rossum distance.

关键词： Encoding Quantization (signal) Neuromorphics Synapses Organizations

Survey on the application of blockchain to IoT 18

学校读者我要写书评

暂无评论

Survey on the application of blockchain to IoT

18th International Conference on Electronics, Information, and Communication, ICEIC 2019

作者： Cho, Sunghyun Lee, Sejong Department of Electrical and Computer Engineering UC San Diego Hanyang University Ansan Korea Republic of Department of Computer Science and Engineering Hanyang University Ansan Korea Republic of

ISBN: (纸本)9788995004449

A network composed of lightweight devices such as IoT has problems due to limited resources such as lack of storage space and low computing performance. These issues pose significant challenges in the application of robust security technologies, which reduces network security performance. The blockchain with strong security is a suitable technology to solve IoT problems with weak security. As a result, various research is being carried out to increase security, lightness, and efficiency of the IoT network by applying blockchain to IoT. This paper introduces the trend of research to apply blockchain to IoT. © 2019 Institute of Electronics and Information Engineers (IEIE).

关键词： Blockchain

Discriminatory or Samaritan - which AI is needed for humanity? An Evolutionary Game Theory Analysis of Hybrid Human-AI populations

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Booker, Tim Miranda, Manuel Moreno López, Jesús A. Fernández, José María Ramos Reddel, Max Widler, Valeria Zimmaro, Filippo Antonioni, Alberto Han, The Anh Complexity Science Hub Vienna Austria IFISC UIB CSIC Spain Universidad de La Laguna Spain Department of Emerging Technology Governance International Center for Future Generations Belgium Modeling and Simulation of Complex Processes Zuse Institute Berlin Germany Department of Mathematics University of Bologna Italy Department of Computer Science University of Pisa Italy Department of Mathematics University Carlos III de Madrid | UC3M Spain School of Computing Engineering and Digital Technologies Teesside University United Kingdom

As artificial intelligence (AI) systems are increasingly embedded in our lives, their presence leads to interactions that shape our behaviour, decision-making, and social interactions. Existing theoretical research has primarily focused on human-to-human interactions, overlooking the unique dynamics triggered by the presence of AI. In this paper, resorting to methods from evolutionary game theory, we study how different forms of AI influence the evolution of cooperation in a human population playing the one-shot Prisoner’s Dilemma game in both well-mixed and structured populations. We found that Samaritan AI agents that help everyone unconditionally, including defectors, can promote higher levels of cooperation in humans than Discriminatory AI that only help those considered worthy/cooperative, especially in slow-moving societies where change is viewed with caution or resistance (small intensities of selection). Intuitively, in fast-moving societies (high intensities of selection), Discriminatory AIs promote higher levels of cooperation than Samaritan AIs. © 2023, CC BY.

关键词： Decision making

A blockchain for media: Survey 18

学校读者我要写书评

暂无评论

A blockchain for media: Survey

18th International Conference on Electronics, Information, and Communication, ICEIC 2019

作者： Cho, Sunghyun Jeong, Chiyoung Electrical and Computer Engineering Division of Software UC San Diego Hanyang University Ansan Korea Republic of Computer Science and Engineering Hanyang University Ansan Korea Republic of

ISBN: (纸本)9788995004449

As the Internet has developed, media distribution services have spread. However, these distribution services have many problems. The blockchain technology can solve these problems. In this paper, this blockchain is defined as media blockchain and is classified into two categories. One is content security that means securing content using blockchain. The other is content right information that means tracing and managing right information using blockchain. We introduce and analyze the media blockchain. © 2019 Institute of Electronics and Information Engineers (IEIE).

关键词： Blockchain

Data Processing of Functional Optical Microscopy for Neuroscience

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Benisty, Hadas Song, Alexander Mishne, Gal Charles, Adam S. Yale Neuroscience New HavenCT06510 United States Max Planck Institute for Intelligent Systems Stuttgart Germany Halıcıoğlu Data Science Institute Department of Electrical and Computer Engineering Neurosciences Graduate Program UC San Diego 9500 Gilman Drive La Jolla CA92093 United States Department of Biomedical Engineering Kavli Neuroscience Discovery Institute Center for Imaging Science Department of Neuroscience Mathematical Institute for Data Science Johns Hopkins University BaltimoreMD21287 United States

Functional optical imaging in neuroscience is rapidly growing with the development of new optical systems and fluorescence indicators. To realize the potential of these massive spatiotemporal datasets for relating neuronal activity to behavior and stimuli and uncovering local circuits in the brain, accurate automated processing is increasingly essential. In this review, we cover recent computational developments in the full data processing pipeline of functional optical microscopy for neuroscience data and discuss ongoing and emerging challenges. © 2022, CC BY-NC-ND.

关键词： Fluorescence microscopy

Robust algorithms for TSP and Steiner tree

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ganesh, Arun Maggs, Bruce M. Panigrahi, Debmalya Department of Electrical Engineering and Computer Sciences UC Berkeley United States Department of Computer Science Duke University Emerald Innovations United States Department of Computer Science Duke University United States

Robust optimization is a widely studied area in operations research, where the algorithm takes as input a range of values and outputs a single solution that performs well for the entire range. Specifically, a robust algorithm aims to minimize regret, defined as the maximum difference between the solution’s cost and that of an optimal solution in hindsight once the input has been realized. For graph problems in P, such as shortest path and minimum spanning tree, robust polynomial-time algorithms that obtain a constant approximation on regret are known. In this paper, we study robust algorithms for minimizing regret in NP-hard graph optimization problems, and give constant approximations on regret for the classical traveling salesman and Steiner tree problems. Copyright © 2020, The Authors. All rights reserved.

关键词： Operations research

Better GPU Hash Tables

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Awad, Muhammad A. Ashkiani, Saman Porumbescu, Serban D. Farach-Colton, Martín Owens, John D. Dept. of Elect. and Comp. Engineering UC Davis DavisCA United States Google Mountain ViewCA United States Dept. of Computer Science Rutgers University PiscatawayNJ United States

We revisit the problem of building static hash tables on the GPU and design and build three bucketed hash tables that use different probing schemes. Our implementations are lock-free and offer efficient memory access patterns;thus, only the probing scheme is the factor affecting the performance of the hash table's different operations. Our results show that a bucketed cuckoo hash table that uses three hash functions (BCHT) outperforms alternative methods that use power-of-two choices, iceberg hashing, and a cuckoo hash table that uses a bucket size one. At high load factors as high as 0.99, BCHT enjoys an average probe count of 1.43 during insertion. Using three hash functions only, positive and negative queries require at most 1.39 and 2.8 average probes per key, respectively. © 2021, CC BY-NC-SA.

关键词： Graphics processing unit

STEM-07. CONSERVATION AND FAITHFUL REPRESENTATION OF CIRCULAR EXTRACHROMOSOMAL DNA IN ORTHOTOPIC PATIENT-DERIVED MEDULLOBLASTOMA XENOGRAFTS

学校读者我要写书评

暂无评论

Neuro-Oncology 2024年第SUPPLEMENT_4期26卷 0–0页

作者： Rishaan Kenkre Owen Chapman Jens Luebeck Vineet Bafna Robert J Wechsler-Reya Jill Mesirov Lukas Chavez Sanford Burnham Prebys Medical Discovery Institute San Diego CA USA UC San Diego Computer Science and Engineering Department San Diego CA USA Columbia University Irving Medical Center New York City NY USA UC San Diego Division of Medical Genetics San Diego CA USA

BACKGROUND Extrachromosomal DNA (ecDNA) drives tumor heterogeneity and correlates with poor survival across various cancer types. Patient-derived xenograft (PDX) models facilitate preclinical drug testing in vivo, enabling tumor growth analysis and molecular tumor profiling. These models, combined with single-cell profiling, offer insights into intratumoral heterogeneity under therapeutic pressure and the emergence of drug resistance. We investigate the abundance and conservation of ecDNA in PDX models to assess the faithful representation of ecDNA compared to their corresponding primary tumors.

关键词：

Memory organization for energy-efficient learning and inference in digital neuromorphic accelerators

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Schaefer, Clemens J.S. Faley, Patrick Neftci, Emre O. Joshi, Siddharth Department of Computer Science and Engineering University of Notre Dame Notre DameIN United States Department of Cognitive Sciences Department of Computer Science UC Irvine IrvineCA United States

The energy efficiency of neuromorphic hardware is greatly affected by the energy of storing, accessing, and updating synaptic parameters. Various methods of memory organisation targeting energy-efficient digital accelerators have been investigated in the past, however, they do not completely encapsulate the energy costs at a system level. To address this shortcoming and to account for various overheads, we synthesize the controller and memory for different encoding schemes and extract the energy costs from these synthesized blocks. Additionally, we introduce functional encoding for structured connectivity such as the connectivity in convolutional layers. Functional encoding offers a 58% reduction in the energy to implement a backward pass and weight update in such layers compared to existing index-based solutions. We show that for a 2 layer spiking neural network trained to retain a spatio-temporal pattern, bitmap (PBBMP) based organization can encode the sparser networks more efficiently. This form of encoding delivers a 1.37× improvement in energy efficiency coming at the cost of a 4% degradation in network retention accuracy as measured by the van Rossum distance. Copyright © 2020, The Authors. All rights reserved.

关键词： Signal encoding