Published on Thu Dec 03 2020

Evolving Character-Level DenseNet Architectures using Genetic Programming

Trevor Londt, Xiaoying Gao, Peter Andreae

It is not clear what DenseNet architectures are optimal for text classification tasks. The iterative task of designing, training and testing of char-DenseNets is an NP-Hard problem that requires expert domain knowledge. The algorithm is evaluated on two popular text datasets.

0
0
0
Abstract

DenseNet architectures have demonstrated impressive performance in image classification tasks, but limited research has been conducted on using character-level DenseNet (char-DenseNet) architectures for text classification tasks. It is not clear what DenseNet architectures are optimal for text classification tasks. The iterative task of designing, training and testing of char-DenseNets is an NP-Hard problem that requires expert domain knowledge. Evolutionary deep learning (EDL) has been used to automatically design CNN architectures for the image classification domain, thereby mitigating the need for expert domain knowledge. This study demonstrates the first work on using EDL to evolve char-DenseNet architectures for text classification tasks. A novel genetic programming-based algorithm (GP-Dense) coupled with an indirect-encoding scheme, facilitates the evolution of performant char DenseNet architectures. The algorithm is evaluated on two popular text datasets, and the best-evolved models are benchmarked against four current state-of-the-art character-level CNN and DenseNet models. Results indicate that the algorithm evolves performant models for both datasets that outperform two of the state-of-the-art models in terms of model accuracy and three of the state-of-the-art models in terms of parameter size.

Thu Dec 03 2020
Neural Networks
Evolving Character-level Convolutional Neural Networks for Text Classification
Character-level convolutional neural networks (char-CNN) require no knowledge of the semantic or syntactic structure of the language they classify. This property simplifies its implementation but reduces its classification accuracy. Research has not established which char-CNN architectures are optimal for text classification tasks.
0
0
0
Wed May 27 2020
Neural Networks
Evolutionary NAS with Gene Expression Programming of Cellular Encoding
The renaissance of neural architecture search (NAS) has seen genetic algorithms (GA) and genetic programming (GP) being used for convolutional neural network (CNN) architectures. While recent work has achieved promising performance on visual perception tasks, the direct encoding scheme of both GA and GP has
0
0
0
Sat Nov 24 2018
Neural Networks
Evolutionary-Neural Hybrid Agents for Architecture Search
Deep Reinforcement Learning based agents can learn complex patterns and explore a vast and compositional search space. The Evo-NAS agent surpasses the accuracy achieved by commonly used agents with only 1/3 of the search cost.
0
0
0
Thu Jan 04 2018
Neural Networks
DENSER: Deep Evolutionary Network Structured Representation
Deep Evolutionary Network Structured Representation (DENSER) is a novel approach to automatically design Artificial Neural Networks (ANNs) using Evolutionary Computation. The algorithm not only searches for the best network topology (e.g., number of layers, type of layers),
0
0
0
Sat Mar 04 2017
Computer Vision
Genetic CNN
The deep Convolutional Neural Network (CNN) is the state-of-the-art solution for large-scale visual recognition. Researchers have manually designed a lot of fixed network structures and verified their effectiveness. In this paper, we discuss the possibility of learning deep network structuresautomatically.
0
0
0
Sun Jul 01 2018
Neural Networks
Autonomous Deep Learning: A Genetic DCNN Designer for Image Classification
Deep convolutional networks (DCNNs) are used in image classification and other vision applications. DCNNs still require a handcrafted design of their architectures. An autonomous learning algorithm can generate a DCNN automatically based on the data available for a specific image.
0
0
0