AutoGO: automated computation graph optimization for neural network evolution

M Salameh, K Mills, N Hassanpour… - Advances in …, 2024 - proceedings.neurips.cc
Abstract Optimizing Deep Neural Networks (DNNs) to obtain high-quality models for efficient
real-world deployment has posed multi-faceted challenges to machine learning engineers …

Activation Space Selectable Kolmogorov-Arnold Networks

Z Yang, J Zhang, X Luo, Z Lu, L Shen - arXiv preprint arXiv:2408.08338, 2024 - arxiv.org
The multilayer perceptron (MLP), a fundamental paradigm in current artificial intelligence, is
widely applied in fields such as computer vision and natural language processing. However …

Building Optimal Neural Architectures using Interpretable Knowledge

KG Mills, FX Han, M Salameh, S Lu… - Proceedings of the …, 2024 - openaccess.thecvf.com
Abstract Neural Architecture Search is a costly practice. The fact that a search space can
span a vast number of design choices with each architecture evaluation taking nontrivial …

How predictors affect the RL-based search strategy in Neural Architecture Search?

J Wu, T Deng, Q Hu - Expert Systems with Applications, 2024 - Elsevier
Abstract Predictor-based Neural Architecture Search is an important topic since it can
efficiently reduce the computational cost of evaluating candidate architectures. Most existing …

Driver fatigue detection method based on multi-feature empirical fusion model

Y Qin, H Lyu, K Zhu - Multimedia Tools and Applications, 2024 - Springer
As the number of long-distance commuters continues to rise, driver fatigue has become a
major contributor to traffic accidents, underscoring the critical need for real-time fatigue …

Task Adaptation of Reinforcement Learning-based NAS Agents through Transfer Learning

A Cassimon, S Mercelis, K Mets - arXiv preprint arXiv:2412.01420, 2024 - arxiv.org
Recently, a novel paradigm has been proposed for reinforcement learning-based NAS
agents, that revolves around the incremental improvement of a given architecture. We …

BSS-Bench: Towards Reproducible and Effective Band Selection Search

W Xu, Z Xu - arXiv preprint arXiv:2312.14570, 2023 - arxiv.org
The key technology to overcome the drawbacks of hyperspectral imaging (expensive, high
capture delay, and low spatial resolution) and make it widely applicable is to select only a …