Additive regularization schedule for neural architecture search
Neural network structures have a critical impact on the accuracy and stability of forecasting. Neural architecture search procedures help design an optimal neural network according to some loss function, which represents a set of quality criteria. This paper investigates the problem of neural network structure optimization. It proposes a way to construct a loss function, which contains a set of additive elements. Each element is called the regularizer. It corresponds to some part of the neural network structure and represents a criterion to optimize. The optimization procedure changes the structure in iterations. To optimize various parts of the structure, the procedure changes the set of regularizers according to some schedule. The authors propose a way to construct the additive regularization schedule. By comparing regularized models with non-regularized ones for a collection of datasets the computational experiments show that the proposed method finds efficient neural network structure and delivers accurate networks of low complexity.
Code (0)
등록된 구현이 없습니다.
Tasks
Efficient Neural NetworkNeural Architecture SearchMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
DARTS-PRIME: Regularization and Scheduling Improve Constrained Optimization in Differentiable NAS
Differentiable Architecture Search (DARTS) is a recent neural architecture search (NAS) method based on a differentiable relaxation. Due to its success, numerous variants analyzing and improving parts of the DARTS framew…
Bilevel OptimizationNeural Architecture SearchSchedulingGenerating Mandarin and Cantonese F0 Contours with Decision Trees and BLSTMs
This paper models the fundamental frequency contours on both Mandarin and Cantonese speech with decision trees and DNNs (deep neural networks). Different kinds of f0 representations and model architectures are tested for…
Rethinking Skip Connections: Additive U-Net for Robust and Interpretable Denoising
Skip connections are central to U-Net architectures for image denoising, but standard concatenation doubles channel dimensionality and obscures information flow, allowing uncontrolled noise transfer. We propose the Addit…
Image DenoisingSCALE: Scalable Cross-Attention Learning with Extrapolation for Agentic Workflow Scheduling
Agentic Large Language Model (LLM) systems decompose complex tasks into workflow Directed Acyclic Graphs (DAGs) whose primitives must be scheduled on heterogeneous clusters. Existing deep reinforcement learning (DRL) sch…
Reinforcement LearningLearning Rate Grafting: Transferability of Optimizer Tuning
In the empirical science of training large neural networks, the learning rate schedule is a notoriously challenging-to-tune hyperparameter, which can depend on all other properties (architecture, optimizer, batch size, d…