4 papers
Electrolyte Bonding Engineering for Highly Uniform GeTe-based CBRAM and Parallel Hebbian Learning in Selector-free Hopfield Networks
Jiin Bang, Jingyeong Hwang, Unhyeon Kang +10
Hopfield networks offer a hardware-friendly framework for energy-efficient associative memory, yet their practical realization in memristor crossbar arrays is critically hindered b…
A2SG:Adaptive and Asymmetric Surrogate Gradients for Training Deep Spiking Neural Networks
Yechan Kang, Yongjin Kweon, Mingyeong Seo +8
Training deep spiking neural networks (SNNs) remains challenging due to sharp loss landscapes and temporal inconsistency caused by surrogate gradients. To address these challenges,…
A Hardware-aware Hopfield Network with a Nonlinear Memristor Array for Robust Associative Memory with Superlinear Capacity
Younghyun Lee, Hakseung Rhee, Unhyeon Kang +9
Associative memory retrieves complete patterns from partial or corrupted inputs and constitutes a primitive form of generative inference. Classical Hopfield networks (CHN) provide…
Neuronal arithmetic operators based on Ovonic threshold switches (OTS) for biologically inspired analog computing
Jingyeong Hwang, Jaesang Lee, Jiin Bang +10
Biological neurons perform arithmetic computations - including additive integration and divisive gain modulation - through synaptic conductance changes and shunting inhibition, ena…