Overview
In nature, brains are built through a process of biological development in which many aspects of the network of neurons and connections change are shaped by external information received through sensory organs. Biological development mechanisms such as axon guidance and dendrite pruning have been shown to rely on neural activity. Despite this, most artificial neural network (ANN) models do not include developmental mechanisms and regard learning as the adjustment of connection weights, while some that do use development restrain it to a period before the ANN is used. It is worthwhile to understand the cognitive functions offered by development and to investigate the fundamental questions raised by artificial neural development. In this workshop, we will explore existing and future approaches that aim to incorporate development into ANNs. Invited speakers will present their work with neural networks, both artificial and biological, in the context of development. Accepted submissions on contemporary work in this field will be presented and we will hold an open discussion on the topic.
2020 Workshop
The 3rd edition of the Developmental Neural Networks Workshop will be held as a part of ALIFE 2020 in its virtual format. To participate and for more information, please see the 2020 workshop page.
Program
10:00-11:30 EDT | Live viewing of presentations | |||
11:30-11:45 EDT | Q&A for Keynote presentation Neuromorphic computers, brains and development: unifying principles and distinctive differences | Simon Laughlin | video | |
11:45-11:55 EDT | Q&A for Developmental neuronal networks as models to study the evolution of biological intelligence | Arend Hintze, P. Robin Hiesinger, Jory Schossau | video | |
11:55-12:05 EDT | Q&A for Multiple problem solving with evolved developmental neural networks: activity dependence | Julian Francis Miller | video | |
12:05-12:15 EDT | Q&A for Developmental Embodied Agents as Meta-brain Models | Bradly Alicea et al. | video | |
12:15-13:00 EDT | Panel discussion on developmental neural networks |
Submissions
We welcome anyone to participate and encourage the submission of extended abstracts of 2-4 pages (MIT Press format). We are open to submissions concerning a variety of neural development topics, from hand-written rules to evolved systems, and from conceptual work to experimental results. Accepted submissions will be distributed online and presented during the workshop. Abstracts must be submitted via email to devonn@irit.fr by June 15th, 2020 UTC-12. This deadline will not be extended.
Important Dates
- Abstract Submission: June 15, 2020
- Author Notification: July 1, 2020
- Workshop: July 18, 2020
- ALIFE: July 13-18, 2020
Organizers
- Julian F Miller, julian.miller@york.ac.uk
- Sylvain Cussat-Blanc, sylvain.cussat-blanc@irit.fr
- Dennis G Wilson, dennis.wilson@isae-supaero.fr
The organizers can be contacted at devonn@irit.fr.
Posts
Multiple problem solving with evolved developmental neural networks: activity dependence
Neuromorphic computers, brains and development: unifying principles and distinctive differences
Developmental neuronal networks as models to study the evolution of biological intelligence
2020 Workshop on Developmental Neural Networks
Developmental Embodied Agents as Meta-brain Models
2019 Workshop on Developmental Neural Networks
2019 Keynote at the Developmental Neural Networks Workshop
First Workshop on Developmental Neural Networks
subscribe via RSS