Chip Learning

Chip learning focuses on developing and implementing machine learning algorithms directly on hardware, aiming for increased efficiency and reduced power consumption compared to traditional cloud-based approaches. Current research emphasizes energy-efficient spiking neural networks (SNNs) and novel training algorithms like Forward-Forward and variations of gradient descent, often incorporating techniques like quantization and tensor decomposition to optimize model size and speed. This field is significant for enabling on-device AI in resource-constrained environments, such as embedded systems and edge devices, with applications ranging from robotics and neuromorphic computing to implantable medical devices.

Papers

October 11, 2024

On-Chip Learning via Transformer In-Context Learning
Jan Finkbeiner, Emre Neftci
Chip Learning Encoder Only Transformer Better on Chip Data Transformer Based in Context

September 19, 2024

The Robustness of Spiking Neural Networks in Communication and its Application towards Network Efficiency in Federated Learning
Manh V. Nguyen, Liang Zhao, Bobin Deng, William Severa, Honghui Xu, Shaoen Wu
Neural Network Native Robustness Federated Learning Spiking Neural Network Timely Communication Chip Learning Efficient Network

August 27, 2024

Distance-Forward Learning: Enhancing the Forward-Forward Algorithm Towards High-Performance On-Chip Learning
Yujie Wu, Siyuan Xu, Jibin Wu, Lei Deng, Mingkun Xu, Qinghao Wen, Guoqi Li
Back Propagation Forward Forward Algorithm Chip Learning Forward Forward Learning

May 1, 2024

Gradient-based Automatic Mixed Precision Quantization for Neural Networks On-Chip
Chang Sun, Thea K. Årrestad, Vladimir Loncar, Jennifer Ngadiuba, Maria Spiropulu
Multiplier Free Quantization Quantization Aware Training Mixed Precision Quantization Uniform Quantization Chip Learning Quantization Granularity

March 30, 2024

Continual Learning for Autonomous Robots: A Prototype-based Approach
Elvin Hajizada, Balachandran Swaminathan, Yulia Sandamirskaya
Continual LEArning Autonomous Robot Prototype Based Prototype Selection Chip Learning Shot Continual Learning

March 5, 2024

DOCTOR: Dynamic On-Chip Temporal Variation Remediation Toward Self-Corrected Photonic Tensor Accelerators
Haotian Lu, Sanmitra Banerjee, Jiaqi Gu
Internal Medicine DOCTOR Photonic Accelerator Chip Learning Photonic Computing Noise Aware Training

December 29, 2023

Darwin3: A large-scale neuromorphic chip with a Novel ISA and On-Chip Learning
De Ma, Xiaofei Jin, Shichun Sun, Yitao Li, Xundong Wu, Youneng Hu, Fangchao Yang, Huajin Tang, Xiaolei Zhu, Peng Lin, Gang Pan
Neural Network Neuromorphic Chip Novel Architecture Chip Learning Large Scale Neuromorphic

October 11, 2023

An On-Chip Trainable Neuron Circuit for SFQ-Based Spiking Neural Networks
Beyza Zeynep Ucpinar, Mustafa Altay Karamuftuoglu, Sasan Razmkhah, Massoud Pedram
Spiking Neural Network Spike Based Chip Learning Neural Network Structure Neuron Attribution

July 12, 2023

Neuromorphic analog circuits for robust on-chip always-on learning in spiking neural networks
Arianna Rubino, Matteo Cartiglia, Melika Payvand, Giacomo Indiveri
Neural Network Chip Design Complementary Metal Oxide Semiconductor Chip Learning Mixed Signal Neuromorphic

June 8, 2023

Mixed-TD: Efficient Neural Network Accelerator with Layer-Specific Tensor Decomposition
Zhewen Yu, Christos-Savvas Bouganis
State of the Art Tensor Decomposition Low Rank Tensor Mixed Speech Chip Learning Dataflow Accelerator

March 5, 2023

Multiplexed gradient descent: Fast online training of modern datasets on hardware neural networks without backpropagation
Adam N. McCaughan, Bakhrom G. Oripov, Natesh Ganesh, Sae Woo Nam, Andrew Dienstfrey, Sonia M. Buckley
Neural Network Gradient Descent Back Propagation Faster Training New Datasets Chip Learning Digital Back Propagation

February 26, 2023

AutoML for neuromorphic computing and application-driven co-design: asynchronous, massively parallel optimization of spiking architectures
Angel Yanguas-Gil, Sandeep Madireddy
Neuromorphic Computing Online autoML Co Design Architecture Perspective Neuromorphic Architecture Chip Learning Complex Structure Parallel Optimization

December 7, 2022

FPGA Implementation of Multi-Layer Machine Learning Equalizer with On-Chip Training
Keren Liu, Erik Börjeson, Christian Häger, Per Larsson-Edefors
Parametric Equalizer Chip Learning FPGA Implementation Neural Network Based Equalizer Adaptive Machine

October 24, 2022

Unlocking the potential of two-point cells for energy-efficient and resilient training of deep nets
Ahsan Adeel, Adewale Adetomi, Khubaib Ahmed, Amir Hussain, Tughrul Arslan, W. A. Phillips
Deep Network Full Potential Neuromorphic Computing Rectified Linear Unit Chip Learning Neuromorphic Artificial Intelligence Architecture Cell Multisensory AI

August 28, 2022

Scalable Nanophotonic-Electronic Spiking Neural Networks
Luis El Srouji, Yun-Jhu Lee, Mehmet Berkay On, Li Zhang, S. J. Ben Yoo
Spiking Neural Network Photonic Device Photonic Neural Network Synaptic Connectivity Chip Learning Silicon Photonics

April 4, 2022

Challenges and Opportunities of Edge AI for Next-Generation Implantable BMIs
MohammadAli Shaeri, Arshia Afzal, Mahsa Shoaran
Technical Challenge Emerging Opportunity Neuroscience Research Neural Signal Chip Learning Brain Machine Interface Edge Artificial Intelligence

November 12, 2021

Silicon Photonic Architecture for Training Deep Neural Networks with Direct Feedback Alignment
Matthew J. Filipovich, Zhimu Guo, Mohammed Al-Qadasi, Bicky A. Marquez, Hugh D. Morison, Volker J. Sorger, Paul R. Prucnal, Sudip Shekhar, Bhavin J. Shastri
Training Deep Feedback Alignment Chip Learning Silicon Photonics Photonic Circuit Fast Neural Network Photonic Processor

Chip Learning

Papers

On-Chip Learning via Transformer In-Context Learning

The Robustness of Spiking Neural Networks in Communication and its Application towards Network Efficiency in Federated Learning

Distance-Forward Learning: Enhancing the Forward-Forward Algorithm Towards High-Performance On-Chip Learning

Gradient-based Automatic Mixed Precision Quantization for Neural Networks On-Chip

Continual Learning for Autonomous Robots: A Prototype-based Approach

DOCTOR: Dynamic On-Chip Temporal Variation Remediation Toward Self-Corrected Photonic Tensor Accelerators

Darwin3: A large-scale neuromorphic chip with a Novel ISA and On-Chip Learning

An On-Chip Trainable Neuron Circuit for SFQ-Based Spiking Neural Networks

Neuromorphic analog circuits for robust on-chip always-on learning in spiking neural networks

Mixed-TD: Efficient Neural Network Accelerator with Layer-Specific Tensor Decomposition

Multiplexed gradient descent: Fast online training of modern datasets on hardware neural networks without backpropagation

AutoML for neuromorphic computing and application-driven co-design: asynchronous, massively parallel optimization of spiking architectures

FPGA Implementation of Multi-Layer Machine Learning Equalizer with On-Chip Training

Unlocking the potential of two-point cells for energy-efficient and resilient training of deep nets

Scalable Nanophotonic-Electronic Spiking Neural Networks

Challenges and Opportunities of Edge AI for Next-Generation Implantable BMIs

Silicon Photonic Architecture for Training Deep Neural Networks with Direct Feedback Alignment