pith. sign in

arxiv: 1703.00548 · v2 · pith:U3ZER2QEnew · submitted 2017-03-01 · 💻 cs.NE · cs.AI

Evolving Deep Neural Networks

Risto Miikkulainen , Jason Liang , Elliot Meyerson , Aditya Rawal , Dan Fink , Olivier Francon , Bala Raju , Hormoz Shahrzad
show 3 more authors
Arshak Navruzyan Nigel Duffy Babak Hodjat
This is my paper
classification 💻 cs.NE cs.AI
keywords deeplearningarchitecturesautomatedevolutionmethodnetworksachieves
0
0 comments X
read the original abstract

The success of deep learning depends on finding an architecture to fit the task. As deep learning has scaled up to more challenging tasks, the architectures have become difficult to design by hand. This paper proposes an automated method, CoDeepNEAT, for optimizing deep learning architectures through evolution. By extending existing neuroevolution methods to topology, components, and hyperparameters, this method achieves results comparable to best human designs in standard benchmarks in object recognition and language modeling. It also supports building a real-world application of automated image captioning on a magazine website. Given the anticipated increases in available computing power, evolution of deep networks is promising approach to constructing deep learning applications in the future.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Evolutionary fine tuning of quantized convolution-based deep learning models

    cs.LG 2026-04 unverdicted novelty 5.0

    Evolutionary fine-tuning of select weights in pre-quantized convolutional networks improves accuracy over standard rounding for VGG, ResNet, and autoencoder models.

  2. Self-Adaptive 2D-3D Ensemble of Fully Convolutional Networks for Medical Image Segmentation

    eess.IV 2019-07 unverdicted novelty 4.0

    Self-adaptive 2D-3D FCN ensemble optimized by multiobjective evolution for prostate segmentation on PROMISE12 achieves top-10 ranking with smaller size than prior auto-designed models.

  3. Genetic Deep Learning for Lung Cancer Screening

    cs.CV 2019-07 unverdicted novelty 3.0

    Genetic algorithm designs a CNN for lung cancer detection in CXRs achieving 97.15% accuracy, outperforming Inception-V3 and ResNet-152 with 4x and 14x fewer parameters.