paper-with-me

홈 › Papers

FENAS: Flexible and Expressive Neural Architecture Search

2020-11-01 · Findings of the Association for Computational Linguistics 2020 · Ramakanth Pasunuru, Mohit Bansal

Architecture search is the automatic process of designing the model or cell structure that is optimal for the given dataset or task. Recently, this approach has shown good improvements in terms of performance (tested on language modeling and image classification) with reasonable training speed using a weight sharing-based approach called Efficient Neural Architecture Search (ENAS). In this work, we propose a novel architecture search algorithm called Flexible and Expressible Neural Architecture Search (FENAS), with more flexible and expressible search space than ENAS, in terms of more activation functions, input edges, and atomic operations. Also, our FENAS approach is able to reproduce the well-known LSTM and GRU architectures (unlike ENAS), and is also able to initialize with them for finding architectures more efficiently. We explore this extended search space via evolutionary search and show that FENAS performs significantly better on several popular text classification tasks and performs similar to ENAS on standard language model benchmark. Further, we present ablations and analyses on our FENAS approach.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

image-classificationImage ClassificationLanguage ModelingLanguage ModellingNeural Architecture Searchtext-classificationText Classification

Methods 이 논문이 사용한 방법론

Tanh Activation 설명 없음
Sigmoid Activation 설명 없음
LSTM An LSTM is a type of recurrent neural network that addresses the vanishing gradient problem in vanilla…
GRU A Gated Recurrent Unit, or GRU, is a type of recurrent neural network. It is similar to an LSTM, but only has two gates - a reset…

Similar Papers 제목 키워드 기반

Accelerating Evolutionary Neural Architecture Search via Multi-Fidelity Evaluation

2021-08-10 · Shangshang Yang, Ye Tian, Xiaoshu Xiang, Shichen Peng 외

Evolutionary neural architecture search (ENAS) has recently received increasing attention by effectively finding high-quality neural architectures, which however consumes high computational cost by training the architect…

GPUNeural Architecture Search

AnchorGT: Efficient and Flexible Attention Architecture for Scalable Graph Transformers

2024-05-06 · Wenhao Zhu, Guojie Song, Liang Wang, Shaoguo Liu

Graph Transformers (GTs) have significantly advanced the field of graph representation learning by overcoming the limitations of message-passing graph neural networks (GNNs) and demonstrating promising performance and ex…

Graph Representation LearningRepresentation Learning

Transferrable Surrogates in Expressive Neural Architecture Search Spaces

2025-04-17 · Shiwen Qin, Gabriela Kadlecová, Martin Pilát, Shay B. Cohen 외

Neural architecture search (NAS) faces a challenge in balancing the exploration of expressive, broad search spaces that enable architectural innovation with the need for efficient evaluation of architectures to effective…

Neural Architecture Search

FAPE: a Constraint-based Planner for Generative and Hierarchical Temporal Planning

2020-10-25 · Arthur Bit-Monnot, Malik Ghallab, Félix Ingrand, David E. Smith

Temporal planning offers numerous advantages when based on an expressive representation. Timelines have been known to provide the required expressiveness but at the cost of search efficiency. We propose here a temporal p…

Homomorphism Counts as Structural Encodings for Graph Learning

2024-10-24 · Linus Bao, Emily Jin, Michael Bronstein, İsmail İlkan Ceylan 외

Graph Transformers are popular neural networks that extend the well-known Transformer architecture to the graph domain. These architectures operate by applying self-attention on graph nodes and incorporating graph struct…

Graph LearningMolecular Property PredictionProperty Prediction