20 results for “analog computing”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
This paper surveys recent advancements in modern analog computing and its applications.
This paper derives a constraint on the fastest normalized mode of analog-computing approaches represented by impedance networks, which is no more than 2π times the largest combined unity-gain bandwidt…
This paper demonstrates that using circuit-level compensation techniques and batch normalization recalibration can mitigate the impact of retention loss on inference accuracy degradation in a single-p…
This paper introduces FerroNDS, a neuromorphic system that uses analog primitives and compute-in-memory hardware to efficiently process neural dynamical systems.
This paper proposes a thermodynamic computing stack for machine learning using stochastic analog processes and energy-based models.
Junyi Yang, Shuai Dong, Zhengnan Fu, Hongyang Shang +1 more
The paper proposes a highly reconfigurable 256x128 in-memory computing array that significantly improves efficiency and performance for analog computing by introducing novel components for ADC, weight…
This paper reduces degradation in memristor-based neural network computations by adjusting ADC settings or removing encoding-related transformations.
A quantized row-height layout synthesis methodology for AMS circuits is proposed to reduce the simulation-to-silicon gap.
AnalogScout is a simulation-efficient PVT-aware sizing framework driven by a large language model agent that reduces power consumption in analog device design.
A hybrid ASIC-FPAA architecture is proposed to enhance security in analog circuits by obfuscating circuit topology using programmable transistor pairs and dummy transistor pairs.
This paper demonstrates input-dependent neuromodulation of a spiking recurrent neural network on a mixed-signal neuromorphic processor.
The authors introduce a new hardware architecture for efficient reinforcement learning using asynchronous Boolean networks.
NEMESIS is a framework that uses large language models to generate accurate performance equations for operational transconductance amplifiers (OTAs) with a balance between speed and accuracy.
Anthony Sánchez, Rafael Ruiz, José A. Belloch, Jesús O. Lacruz +1 more
This paper investigates the use of Spiking Neural Networks (SNNs) for efficient micro-Doppler activity recognition in wireless networks.
This paper proposes a method to use large language models as verification assistants for analog circuits.