6 papers
FAAST: Forward-Only Associative Learning via Closed-Form Fast Weights for Test-Time Supervised Adaptation
Guangsheng Bao, Hongbo Zhang, Han Cui +4
Adapting pretrained models typically involves a trade-off between the high training costs of backpropagation and the heavy inference overhead of memory-based or in-context learning…
Reconfigurable Digital RRAM Logic Enables In-Situ Pruning and Learning for Edge AI
Songqi Wang, Yue Zhang, Jia Chen +9
The human brain simultaneously optimizes synaptic weights and topology by growing, pruning, and strengthening synapses while performing all computation entirely in memory. In contr…
Topology Optimization of Random Memristors for Input-Aware Dynamic SNN
Bo Wang, Shaocong Wang, Ning Lin +12
There is unprecedented development in machine learning, exemplified by recent large language models and world simulators, which are artificial neural networks running on digital co…
Dynamic neural network with memristive CIM and CAM for 2D and 3D vision
Yue Zhang, Woyu Zhang, Shaocong Wang +14
The brain is dynamic, associative and efficient. It reconfigures by associating the inputs with past experiences, with fused memory and processing. In contrast, AI models are stati…
Older and Wiser: The Marriage of Device Aging and Intellectual Property Protection of Deep Neural Networks
Ning Lin, Shaocong Wang, Yue Zhang +6
Deep neural networks (DNNs), such as the widely-used GPT-3 with billions of parameters, are often kept secret due to high training costs and privacy concerns surrounding the data u…
Efficient and accurate neural field reconstruction using resistive memory
Yifei Yu, Shaocong Wang, Woyu Zhang +16
Human beings construct perception of space by integrating sparse observations into massively interconnected synapses and neurons, offering a superior parallelism and efficiency. Re…