Search Results for author:
Found 31 papers, 7 papers with code
Biologically-Inspired Continual Learning of Human Motion Sequences
This work proposes a model for continual learning on tasks involving temporal sequences, specifically, human motions.
Continuous-Time Analog Filters for Audio Edge Intelligence: Review on Circuit Designs
Edge audio devices can reduce data bandwidth requirements by pre-processing input speech on the device before transmission to the cloud.
Optimizing the Consumption of Spiking Neural Networks with Activity Regularization
Reducing energy consumption is a critical point for neural network models running on edge devices.
Kernel Modulation: A Parameter-Efficient Method for Training Convolutional Neural Networks
This work proposes a novel parameter-efficient kernel modulation (KM) method that adapts all parameters of a base network instead of a subset of layers.
Spiking Cochlea with System-level Local Automatic Gain Control
The bandpass filter gain of a channel is adapted dynamically to the input amplitude so that the average output spike rate stays within a defined range.
Exploiting Spatial Sparsity for Event Cameras with Visual Transformers
The input to the ViT consists of events that are accumulated into time bins and spatially separated into non-overlapping sub-regions called patches.
T-NGA: Temporal Network Grafting Algorithm for Learning to Process Spiking Audio Sensor Events
This work proposes a self-supervised method called Temporal Network Grafting Algorithm (T-NGA), which grafts a recurrent network pretrained on spectrogram features so that the network works with the cochlea event features.
Spartus: A 9.4 TOp/s FPGA-based LSTM Accelerator Exploiting Spatio-Temporal Sparsity
The pruned networks running on Spartus hardware achieve weight sparsity levels of up to 96% and 94% with negligible accuracy loss on the TIMIT and the Librispeech datasets.
Prospects for Analog Circuits in Deep Networks
Emerging dense non-volatile memory technologies can help to provide on-chip memory and analog circuits can be well suited to implement the needed multiplication-vector operations coupled with in-computing memory approaches.
v2e: From Video Frames to Realistic DVS Events
The first experiment is object recognition with N-Caltech 101 dataset.
DDD20 End-to-End Event Camera Driving Dataset: Fusing Frames and Events with Deep Learning for Improved Steering Prediction
The dataset was captured with a DAVIS camera that concurrently streams both dynamic vision sensor (DVS) brightness change events and active pixel sensor (APS) intensity frames.
Data-Driven Neuromorphic DRAM-based CNN and RNN Accelerators
The energy consumed by running large deep neural networks (DNNs) on hardware accelerators is dominated by the need for lots of fast memory to store both states and weights.
Learning to Exploit Multiple Vision Modalities by Using Grafted Networks
This paper proposes a Network Grafting Algorithm (NGA), where a new front end network driven by unconventional visual inputs replaces the front end network of a pretrained deep network that processes intensity frames.
Recurrent Neural Network Control of a Hybrid Dynamic Transfemoral Prosthesis with EdgeDRNN Accelerator
Lower leg prostheses could improve the life quality of amputees by increasing comfort and reducing energy to locomote, but currently control methods are limited in modulating behaviors based upon the human's experience.
EdgeDRNN: Enabling Low-latency Recurrent Neural Network Edge Inference
This paper presents a Gated Recurrent Unit (GRU) based recurrent neural network (RNN) accelerator called EdgeDRNN designed for portable edge computing.
FaSNet: Low-latency Adaptive Beamforming for Multi-microphone Audio Processing
Beamforming has been extensively investigated for multi-channel audio processing tasks.
Closing the Accuracy Gap in an Event-Based Visual Recognition Task
Mobile and embedded applications require neural networks-based pattern recognition systems to perform well under a tight computational budget.
Reducing state updates via Gaussian-gated LSTMs
Finally, by employing a temporal curriculum learning schedule for the g-LSTM, we can reduce the convergence time of the equivalent LSTM network on long sequences.
Gaussian-gated LSTM: Improved convergence by reducing state updates
Finally, by employing a temporal curriculum learning schedule for the g-LSTM, we can reduce the convergence time of the equivalent LSTM network on long sequences.
Overcoming the vanishing gradient problem in plain recurrent networks
Plain recurrent networks greatly suffer from the vanishing gradient problem while Gated Neural Networks (GNNs) such as Long-short Term Memory (LSTM) and Gated Recurrent Unit (GRU) deliver promising results in many sequence learning tasks through sophisticated network designs.
DDD17: End-To-End DAVIS Driving Dataset
Event cameras, such as dynamic vision sensors (DVS), and dynamic and active-pixel vision sensors (DAVIS) can supplement other autonomous driving sensors by providing a concurrent stream of standard active pixel sensor (APS) images and DVS temporal contrast events.
Sensor Transformation Attention Networks
Recent work on encoder-decoder models for sequence-to-sequence mapping has shown that integrating both temporal and spatial attention mechanisms into neural networks increases the performance of the system substantially.
NullHop: A Flexible Convolutional Neural Network Accelerator Based on Sparse Representations of Feature Maps
By exploiting sparsity, NullHop achieves an efficiency of 368%, maintains over 98% utilization of the MAC units, and achieves a power efficiency of over 3TOp/s/W in a core area of 6. 3mm$^2$.
Delta Networks for Optimized Recurrent Network Computation
Similarly, on the large Wall Street Journal speech recognition benchmark even existing networks can be greatly accelerated as delta networks, and a 5. 7x improvement with negligible loss of accuracy can be obtained through training.
Recurrent Neural Networks With Limited Numerical Precision
Recurrent Neural Networks (RNNs) produce state-of-art performance on many machine learning tasks but their demand on resources in terms of memory and computational power are often high.
Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences
In this work, we introduce the Phased LSTM model, which extends the LSTM unit by adding a new time gate.
Recurrent Neural Networks With Limited Numerical Precision
We present results from the use of different stochastic and deterministic reduced precision training methods applied to three major RNN types which are then tested on several datasets.
Precise neural network computation with imprecise analog devices
The operations used for neural network computation map favorably onto simple analog circuits, which outshine their digital counterparts in terms of compactness and efficiency.
A Curriculum Learning Method for Improved Noise Robustness in Automatic Speech Recognition
The performance of automatic speech recognition systems under noisy environments still leaves room for improvement.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
Memory and information processing in neuromorphic systems
We describe the advantages of the different approaches being pursued and present the challenges that need to be addressed for building artificial neural processing systems that can display the richness of behaviors seen in biological systems.
Hardware-Amenable Structural Learning for Spike-based Pattern Classification using a Simple Model of Active Dendrites
This work also presents a branch-specific spike-based version of this structural plasticity rule.
