Our paper, “Dynamic Optimization of Neural Network Structures Using Probabilistic Modeling”, has been accepted to the Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18). This paper proposes the framework to dynamically optimize the neural network structure during the training using gradient-based method.

We won the Best Paper Award of Evolutionary Machine Learning (EML) Track in GECCO 2017! The title of the award paper is “A Genetic Programming Approach to Designing Convolutional Neural Network Architectures”. This paper achieved to find high-performance architectures of the convolutional neural networks by the genetic programming with the reasonable computational resource. Masanori Suganuma, Shinichi Shirakawa, and Tomoharu Nagao: A Genetic Programming Approach to Designing Convolutional Neural Network Architectures, Genetic and Evolutionary Computation Conference 2017 (GECCO 2017), pp. [Read More]

The members’ page has been updated. Now, our laboratory has two graduate and six undergraduate students.

Members page

Lecturer Shirakawa gave a research talk in the IEEE World Congress on Computational Intelligence (WCCI) 2016 which was held in Vancouver, Canada, 24-29 July.

The title of the presentation is “Impact of Invariant Objective for Order Preserving Transformation in Bayesian Optimization.”

In April 2016, our laboratory has started with Lecturer Shirakawa and two undergraduate students.