Deep Learning for Human Activity Recognition Based on Causality Feature Extraction by Yu Min Hwang, Sangjun Park, Hyung Ok Lee, Seok-Kap Ko, Byung-Tak Lee

“…Subsequently, the causality and fuzzy features are fused by matrix multiplication to express distinct features of activities. …”
Get full text

Multi-Scale Dense Graph Attention Network for Hyperspectral Classification by Chen Wang, Lu Li, ZhongQi Wang, JingYao Ma, YunLong Kong, YanFeng Wang, JianRui Chang, ZiMeng Zhang, XinYu Lin

“…Subsequently, an improved graph self-attention network is sequentially applied to each scale graph, and the different scale graphs are densely connected through spatial feature alignment modules, designed using twice matrix multiplication. Finally, the combined pixel-level feature map from multiple graph spaces is derived, and Spectral-Spatial convolution is employed to fuse the abundant feature maps for hyperspectral classification. …”
Get full text

STT-BSNN: An In-Memory Deep Binary Spiking Neural Network Based on STT-MRAM by Van-Tinh Nguyen, Quang-Kien Trinh, Renyuan Zhang, Yasuhiko Nakashima

“…When the common inputs are fed through BLs, vector-to-matrix multiplication can be performed in a single memory sensing phase, hence achieving massive parallelism with low power and low latency. …”
Get full text

Shortest paths, Markov chains, matrix scaling and beyond : improved algorithms through the lens of continuous optimization by Vladu, Adrian Valentin

Get full text

Faster deterministic and Las vegas algorithms for offline approximate nearest neighbors in high dimensions by Alman, Joshua, Williams, R Ryan

“…The algorithm follows from a simple combination of expander walks, Chebyshev polynomials, and rectangular matrix multiplication. We also show how to eliminate errors in the previous Monte Carlo randomized algorithm of Alman, Chan, and Williams [FOCS'16] for offline approximate nearest or farthest neighbors, and obtain a Las Vegas randomized algorithm with expected running time n2−Ω(ε1/3/log(1/ε)) . …”
Get full text

Approximating cycles in directed graphs: Fast algorithms for girth and roundtrip spanners by Pachocki, J, Roditty, L, Sidford, A, Tov, R, Williams, Virginia Vassilevska

“…We show that the runtime of our algorithm cannot be significantly improved without a breakthrough in combinatorial boolean matrix multiplication. We also show that if the girth is O(na), then the same guarantee can be achieved via a deterministic algorithm. …”
Get full text

A line feature extraction method for finger-knuckle-print verification by Kim, Jooyoung, Oh, Kangrok, Oh, Beom-Seok, Lin, Zhiping, Toh, Kar-Ann

“…Such a matrix enables directional image shifting and subtraction within a single matrix multiplication. The resultant difference image then goes through a sigmoidal activation for contrast enhancement. …”
Get full text

Jordan-chevalley decomposition of recorded electroencephalography signals during epileptic seizures by Ahmad Fuad, Amirul Aizad

Get full text

A Conv‐GEMM reconfigurable accelerator with WS‐RS dataflow for high throughput processing by Feihu Wang, Chi Zhang, Yongchao Deng, Xu Yang, Shuangming Yu, Runjiang Dou, Nanjian Wu, Liyuan Liu

“…However, the significant differences between convolution and matrix computation patterns have posed a challenge in efficiently supporting both convolution (Conv) and general matrix multiplication (GEMM) on hardware design. This paper proposes a Conv‐GEMM reconfigurable accelerator architecture for high throughput edge processing. …”
Get full text

Fast mutual exclusivity algorithm nominates potential synthetic lethal gene pairs through brute force matrix product computations by Tarcisio Fedrizzi, Yari Ciani, Francesca Lorenzin, Thomas Cantore, Paola Gasperini, Francesca Demichelis

“…Here we present Fast Mutual Exclusivity (FaME), an algorithm based on matrix multiplication that employs a logarithm-based implementation of the Fisher’s exact test to achieve fast computation of genome-wide mutual exclusivity tests; we show that brute force testing for mutual exclusivity of hundreds of millions of aberrations combinations can be performed in few minutes. …”
Get full text

PISA-DMA: Processing-in-Memory Instruction Set Architecture Using DMA by Won Jun Lee, Chang Hyun Kim, Yoonah Paik, Seon Wook Kim

“…Also, we study how our proposed PISA affects performance in compiler optimization and show that the operator fusion of matrix-matrix multiplication and element-wise addition achieved 1.04x speedup, a similar performance gain using conventional ISAs.…”
Get full text

Nonvolatile Capacitive Crossbar Array for In‐Memory Computing by Jae Hur, Yuan-Chun Luo, Anni Lu, Tzu-Han Wang, Shaolan Li, Asif Islam Khan, Shimeng Yu

“…Asymmetry of the HZO electrode interfaces leads to small‐signal capacitance on/off ratio >110% that can achieve read‐disturb‐free operation. The vector matrix multiplication (VMM) experiments are conducted on the fabricated capacitive crossbar array, showing a linear weighted sum versus numbers of input or on‐state weight. …”
Get full text

Speeding up eQTL scans in the BXD population using GPUs by Chelsea Trotter, Hyeonju Kim, Gregory Farage, Pjotr Prins, Robert W Williams, Karl W Broman, Śaunak Sen

“…By using easily parallelizable operations including matrix multiplication, vectorized operations, and element-wise operations, our method is more than 700 times faster than a R/qtl linear model genome scan using 16 threads. …”
Get full text

Performance Evaluation of ARM-based versus x86- based Processors in High Performance Computing Clusters by Dileep Kumar, Pardeep Kumar, Muhammad Awais Rajput, Saleem Ahmed, Muhammad Saqib Bhatti, Anthony Tsetse

“…To compare the performances of both clusters, two complex applications were implemented: approximation of mathematical constant PI (3.14), a matrix multiplication program. Our analysis focused on execution time and speedup as performance metrics. …”
Get full text

A hierarchical multi-criteria model for analyzing the barriers to Pharma 4.0 implementation in developing countries by Akib Zaman, Ismat Jerin, Puja Ghosh, Anika Akther, Salma Sultana Shrity, Ferdous Sarwar

“…Subsequently, we analyzed the correlation and influence of the selected barriers on each other by formulating a hierarchical multi-criteria model integrating Interpretive Structural Modelling (ISM) and the Cross-Impact Matrix Multiplication Applied to Classification (MICMAC). …”
Get full text

DARKSIDE: A Heterogeneous RISC-V Compute Cluster for Extreme-Edge On-Chip DNN Inference and Training by Angelo Garofalo, Yvan Tortorella, Matteo Perotti, Luca Valente, Alessandro Nadalini, Luca Benini, Davide Rossi, Francesco Conti

“…To boost the performance and efficiency on key compute-intensive DNN kernels, the cluster is enriched with three digital accelerators: 1) a specialized engine for low-data-reuse depthwise convolution kernels (up to 30 MAC/cycle); 2) a minimal overhead datamover to marshal 1–32-b data on-the-fly; and 3) a 16-b floating-point tensor product engine (TPE) for tiled matrix-multiplication acceleration. DARKSIDE is implemented in 65-nm CMOS technology. …”
Get full text

TS-Perf: General Performance Measurement of Trusted Execution Environment and Rich Execution Environment on Intel SGX, Arm TrustZone, and RISC-V Keystone by Kuniyasu Suzaki, Kenta Nakajima, Tsukasa Oi, Akira Tsukamoto

“…TS-perf measures the performance of GlobalPlatform’s TEE internal APIs, matrix multiplication, memory access, and storage access. …”
Get full text

PIMCaffe: Functional Evaluation of a Machine Learning Framework for In-Memory Neural Processing Unit by Won Jeon, Jiwon Lee, Dongseok Kang, Hongju Kal, Won Woo Ro

“…Our prototype NPU design includes a vector processor for parallel vector processing and a systolic array unit for matrix multiplication. Using the proposed software framework, we perform a detailed analysis of the in-memory neural processing unit. …”
Get full text

A Compact Model for Interface-Type Self-Rectifying Resistive Memory With Experiment Verification by Jin-Woo Kim, Jun-Seok Beom, Hong-Sub Lee, Nam-Seog Kim

“…A nano-battery bias scheme is also proposed to zero the current of RRAMs having non-zero I-V crossing points, reducing the sneak-pass current error to 0.02%. A vector matrix multiplication application demonstrates 3.44 TOPS/W with a 50:50 LRS to HRS ratio, and a deep neural network on a VGG-8 architecture using the CIFAR-10 dataset observes an accuracy degradation of 1.36%.…”
Get full text

TA-YOLO: a lightweight small object detection model based on multi-dimensional trans-attention module for remote sensing images by Minze Li, Yuling Chen, Tao Zhang, Wu Huang

“…First, in order to reduce computational complexity and improve feature richness, we use a special linear convolution to obtain three projection features instead of the simple matrix multiplication transformation in Transformer. Second, we obtain trans-attention maps in different dimensions in a manner similar to the self-attention mechanism to capture the interrelationships of features in channels and spaces. …”
Get full text