Search Results for author:
Found 15 papers, 8 papers with code
Zero-Cost Proxies for Lightweight NAS
For example, Spearman's rank correlation coefficient between final validation accuracy and our best zero-cost proxy on NAS-Bench-201 is 0. 82, compared to 0. 61 for EcoNAS (a recently proposed reduced-training proxy).
Dynamic Channel Pruning: Feature Boosting and Suppression
Making deep convolutional neural networks more accurate typically comes at the cost of increased computational and memory resources.
$μ$NAS: Constrained Neural Architecture Search for Microcontrollers
IoT devices are powered by microcontroller units (MCUs) which are extremely resource-scarce: a typical MCU may have an underpowered processor and around 64 KB of memory and persistent storage, which is orders of magnitude fewer computational resources than is typically required for deep learning.
BRP-NAS: Prediction-based NAS using GCNs
What is more, we investigate prediction quality on different metrics and show that sample efficiency of the predictor-based NAS can be improved by considering binary relations of models and an iterative data selection strategy.
NAS-Bench-ASR: Reproducible Neural Architecture Search for Speech Recognition
These datasets, however, focus predominantly on computer vision and NLP tasks and thus suffer from the problem of limited coverage of application domains.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
Journey Towards Tiny Perceptual Super-Resolution
Recent works in single-image perceptual super-resolution (SR) have demonstrated unprecedented performance in generating realistic textures by means of deep convolutional networks.
BLOX: Macro Neural Architecture Search Benchmark and Algorithms
To provide a systematic study of the performance of NAS algorithms on a macro search space, we release Blox - a benchmark that consists of 91k unique models trained on the CIFAR-100 dataset.
Neural Fine-Tuning Search for Few-Shot Learning
In few-shot recognition, a classifier that has been trained on one set of classes is required to rapidly adapt and generalize to a disjoint, novel set of classes.
ShrinkML: End-to-End ASR Model Compression Using Reinforcement Learning
Our results show that in the absence of retraining our RL-based search is an effective and practical method to compress a production-grade ASR system.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+6
MobiSR: Efficient On-Device Super-Resolution through Heterogeneous Mobile Processors
In recent years, convolutional networks have demonstrated unprecedented performance in the image restoration task of super-resolution (SR).
Best of Both Worlds: AutoML Codesign of a CNN and its Hardware Accelerator
We automate HW-CNN codesign using NAS by including parameters from both the CNN model and the HW accelerator, and we jointly search for the best model-accelerator pair that boosts accuracy and efficiency.
Iterative Compression of End-to-End ASR Model using AutoML
Increasing demand for on-device Automatic Speech Recognition (ASR) systems has resulted in renewed interests in developing automatic model compression techniques.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
Zero-Cost Operation Scoring in Differentiable Architecture Search
From this perspective, we introduce a novel \textit{perturbation-based zero-cost operation scoring} (Zero-Cost-PT) approach, which utilizes zero-cost proxies that were recently studied in multi-trial NAS but degrade significantly on larger search spaces, typical for differentiable NAS.
Federated Learning for Inference at Anytime and Anywhere
Federated learning has been predominantly concerned with collaborative training of deep networks from scratch, and especially the many challenges that arise, such as communication cost, robustness to heterogeneous data, and support for diverse device capabilities.
How Much Is Hidden in the NAS Benchmarks? Few-Shot Adaptation of a NAS Predictor
Neural architecture search has proven to be a powerful approach to designing and refining neural networks, often boosting their performance and efficiency over manually-designed variations, but comes with computational overhead.
