Shih‐Chii Liu
YOU?
Author Swipe
View article: Deep Learning–Based Control of Electrically Evoked Activity in Human Visual Cortex
Deep Learning–Based Control of Electrically Evoked Activity in Human Visual Cortex Open
Visual cortical prostheses offer a promising path to sight restoration, but current systems elicit crude, variable percepts and rely on manual electrode-by-electrode calibration that does not scale. This work introduces an automated data-d…
View article: Advanced analysis of fully-printed organic transistors platform for multi-ion detection in sweat
Advanced analysis of fully-printed organic transistors platform for multi-ion detection in sweat Open
View article: EFLOP: A Sparsity-Aware Metric for Evaluating Computational Cost in Spiking and Non-Spiking Neural Networks
EFLOP: A Sparsity-Aware Metric for Evaluating Computational Cost in Spiking and Non-Spiking Neural Networks Open
Deploying energy-efficient deep neural networks on energy-constrained edge devices is an important research topic in both machine learning and circuit design communities. Both ANNs and SNNs have been proposed as candidates for these tasks …
View article: The neurobench framework for benchmarking neuromorphic computing algorithms and systems
The neurobench framework for benchmarking neuromorphic computing algorithms and systems Open
Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it diffic…
View article: Real-time control of a hearing instrument with EEG-based attention decoding
Real-time control of a hearing instrument with EEG-based attention decoding Open
Enhancing speech perception in everyday noisy acoustic environments remains an outstanding challenge for hearing aids. Speech separation technology is improving rapidly, but hearing devices cannot fully exploit this advance without knowing…
View article: Modulating State Space Model with SlowFast Framework for Compute-Efficient Ultra Low-Latency Speech Enhancement
Modulating State Space Model with SlowFast Framework for Compute-Efficient Ultra Low-Latency Speech Enhancement Open
Deep learning-based speech enhancement (SE) methods often face significant computational challenges when needing to meet low-latency requirements because of the increased number of frames to be processed. This paper introduces the SlowFast…
View article: Leveraging Recurrent Neural Networks for Predicting Motor Movements from Primate Motor Cortex Neural Recordings
Leveraging Recurrent Neural Networks for Predicting Motor Movements from Primate Motor Cortex Neural Recordings Open
This paper presents an efficient deep learning solution for decoding motor movements from neural recordings in non-human primates. An Autoencoder Gated Recurrent Unit (AEGRU) model was adopted as the model architecture for this task. The a…
View article: Dynamic Gated Recurrent Neural Network for Compute-efficient Speech Enhancement
Dynamic Gated Recurrent Neural Network for Compute-efficient Speech Enhancement Open
This paper introduces a new Dynamic Gated Recurrent Neural Network (DG-RNN)\nfor compute-efficient speech enhancement models running on resource-constrained\nhardware platforms. It leverages the slow evolution characteristic of RNN\nhidden…
View article: Text-to-Events: Synthetic Event Camera Streams from Conditional Text Input
Text-to-Events: Synthetic Event Camera Streams from Conditional Text Input Open
Event cameras are advantageous for tasks that require vision sensors with low-latency and sparse output responses. However, the development of deep network algorithms using event cameras has been slow because of the lack of large labelled …
View article: DeltaKWS: A 65nm 36nJ/Decision Bio-inspired Temporal-Sparsity-Aware Digital Keyword Spotting IC with 0.6V Near-Threshold SRAM
DeltaKWS: A 65nm 36nJ/Decision Bio-inspired Temporal-Sparsity-Aware Digital Keyword Spotting IC with 0.6V Near-Threshold SRAM Open
This paper introduces DeltaKWS, to the best of our knowledge, the first $Δ$RNN-enabled fine-grained temporal sparsity-aware KWS IC for voice-controlled devices. The 65 nm prototype chip features a number of techniques to enhance performanc…
View article: Event-Based Eye Tracking. AIS 2024 Challenge Survey
Event-Based Eye Tracking. AIS 2024 Challenge Survey Open
This survey reviews the AIS 2024 Event-Based Eye Tracking (EET) Challenge. The task of the challenge focuses on processing eye movement recorded with event cameras and predicting the pupil center of the eye. The challenge emphasizes effici…
View article: Exploiting Symmetric Temporally Sparse BPTT for Efficient RNN Training
Exploiting Symmetric Temporally Sparse BPTT for Efficient RNN Training Open
Recurrent Neural Networks (RNNs) are useful in temporal sequence tasks. However, training RNNs involves dense matrix multiplications which require hardware that can support a large number of arithmetic operations and memory accesses. Imple…
View article: Real-time control of a hearing instrument with EEG-based attention decoding
Real-time control of a hearing instrument with EEG-based attention decoding Open
Enhancing speech perception in everyday noisy acoustic environments remains an outstanding challenge for hearing aids. Speech separation technology is improving rapidly, but hearing devices cannot fully exploit this advance without knowing…
View article: Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer
Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer Open
Epilepsy is a common disease of the nervous system. Timely prediction of seizures and intervention treatment can significantly reduce the accidental injury of patients and protect the life and health of patients. This paper presents a neur…
View article: Bringing Dynamic Sparsity to the Forefront for Low-Power Audio Edge Computing: Brain-inspired approach for sparsifying network updates
Bringing Dynamic Sparsity to the Forefront for Low-Power Audio Edge Computing: Brain-inspired approach for sparsifying network updates Open
Dynamic sparsity is intrinsic to biological computing and is key to its extreme power efficiency. Edge computing systems can improve their energy efficiency and reduce response latency by exploiting this neuromorphic principle. The neuromo…
View article: Exploiting Symmetric Temporally Sparse BPTT for Efficient RNN Training
Exploiting Symmetric Temporally Sparse BPTT for Efficient RNN Training Open
Recurrent Neural Networks (RNNs) are useful in temporal sequence tasks. However, training RNNs involves dense matrix multiplications which require hardware that can support a large number of arithmetic operations and memory accesses. Imple…
View article: Guest Editorial Dynamical Neuro-AI Learning Systems: Devices, Circuits, Architecture and Algorithms
Guest Editorial Dynamical Neuro-AI Learning Systems: Devices, Circuits, Architecture and Algorithms Open
This Special Issue of IEEE Journal on Emerging and Selected Topics in Circuits and Systems (JETCAS) is dedicated to demonstrating the latest research progress on dynamical neuro-artificial intelligence (AI) learning systems that bridge the…
View article: A 128-channel real-time VPDNN stimulation system for a visual cortical neuroprosthesis
A 128-channel real-time VPDNN stimulation system for a visual cortical neuroprosthesis Open
With the recent progress in developing large-scale micro-electrodes, cortical neuroprotheses supporting hundreds of electrodes will be viable in the near future. We describe work in building a visual stimulation system that receives camera…
View article: Multisensing Wearables for Real-Time Monitoring of Sweat Electrolyte Biomarkers During Exercise and Analysis on Their Correlation With Core Body Temperature
Multisensing Wearables for Real-Time Monitoring of Sweat Electrolyte Biomarkers During Exercise and Analysis on Their Correlation With Core Body Temperature Open
Sweat secreted by the human eccrine sweat glands can provide valuable biomarker information during exercise. Real-time non-invasive biomarker recordings are therefore useful for evaluating the physiological conditions of an athlete such as…
View article: Analytical assessment of sodium ISFET based sensors for sweat analysis
Analytical assessment of sodium ISFET based sensors for sweat analysis Open
Body thermoregulation during exercise induces sweating with the consequent loss of water, electrolytes and other compounds. The use of sweat for bioanalytical purposes has recently widespread because it is an easily accessible biofluid tha…
View article: High-Accuracy and Energy-Efficient Acoustic Inference using Hardware-Aware Training and a 0.34nW/Ch Full-Wave Rectifier
High-Accuracy and Energy-Efficient Acoustic Inference using Hardware-Aware Training and a 0.34nW/Ch Full-Wave Rectifier Open
A full-wave rectifier (FWR) is a necessary component of many analog acoustic feature extractor (FEx) designs targeted at edge audio applications. However, analog circuits that perform close-to-ideal rectification contribute a significant p…
View article: Biologically-Inspired Continual Learning of Human Motion Sequences
Biologically-Inspired Continual Learning of Human Motion Sequences Open
This work proposes a model for continual learning on tasks involving temporal sequences, specifically, human motions. It improves on a recently proposed brain-inspired replay model (BI-R) by building a biologically-inspired conditional tem…
View article: A 3.11 μW 40 nV/ √Hz Instrumentation Amplifier for Bio-Impedance Sensors Exploiting Positive-Feedback-Assisted Gain Boosting
A 3.11 μW 40 nV/ √Hz Instrumentation Amplifier for Bio-Impedance Sensors Exploiting Positive-Feedback-Assisted Gain Boosting Open
This paper presents a low-power and low-noise instrumentation amplifier (IA) for bio-impedance sensing applications. To solve the loop gain reduction problem when the input resistor in a transconductance (TC) stage decreases as low as <10 …
View article: An Area-Efficient Ultra-Low-Power Time-Domain Feature Extractor for Edge Keyword Spotting
An Area-Efficient Ultra-Low-Power Time-Domain Feature Extractor for Edge Keyword Spotting Open
Keyword spotting (KWS) is an important task on edge low-power audio devices. A typical edge KWS system consists of a front-end feature extractor which outputs mel-scale frequency cepstral coefficients (MFCC) features followed by a back-end…
View article: End-to-End Prediction of Sodium Concentration from Uncalibrated Sodium ISFETs
End-to-End Prediction of Sodium Concentration from Uncalibrated Sodium ISFETs Open
Ion-selective field-effect transistors (ISFETs) are widely used for chemical sensing in biomedical and environmental applications. They require calibration before deployment in the field because of individual sensor response variations and…
View article: Analytical Assessment of Sodium Isfet Based Sensors for Sweat Analysis
Analytical Assessment of Sodium Isfet Based Sensors for Sweat Analysis Open
View article: Energy-efficient activity-driven computing architectures for edge intelligence
Energy-efficient activity-driven computing architectures for edge intelligence Open
We present an overview of different methods to increase the energy efficiency of Tiny Machine Learning networks running on resource-constrained edge neural network accelerators. Most commonly reported are accelerator designs that support n…
View article: Real-time smart multisensing wearable platform for monitoring sweat biomarkers during exercise
Real-time smart multisensing wearable platform for monitoring sweat biomarkers during exercise Open
Sweat secreted by the human eccrine sweat glands can provide valuable biomarker information during exercise in hot and humid conditions. Real-time noninvasive biomarker recordings are therefore useful for evaluating the physiological condi…
View article: Person identification using deep neural networks on physiological biomarkers during exercise
Person identification using deep neural networks on physiological biomarkers during exercise Open
Much progress has been made in wearable sensors that provide real-time continuous physiological data from non- invasive measurements including heart rate and biofluids such as sweat. This information can potentially be used to identify the…
View article: A 23-<i>μ</i>W Keyword Spotting IC With Ring-Oscillator-Based Time-Domain Feature Extraction
A 23-<i>μ</i>W Keyword Spotting IC With Ring-Oscillator-Based Time-Domain Feature Extraction Open
This article presents the first keyword spotting (KWS) IC that uses a ring-oscillator-based time-domain processing technique for its analog feature extractor (FEx). Its extensive usage of time-encoding schemes allows the analog audio signa…