ORCID: https://orcid.org/0009-0006-2750-7495; Lenta, Cristian; Kölle, Michael; Linnhoff-Popien, Claudia
ORCID: https://orcid.org/0000-0001-6284-9286 und Gabor, Thomas
ORCID: https://orcid.org/0000-0003-2048-8667
(2022):
Constructing Organism Networks from Collaborative Self-Replicators.
SSCI 2022: IEEE Symposium Series on Computational Intelligence, Singapore, Singapore, 04. - 07. Dezember 2022.
Ishibuchi, Hisao; Kwoh, Chee-Keong; Tan, Ah-Hwee; Srinivasan, Dipti; Miao, Chunyan; Trivedi, Anupam und Crockett, Keeley (eds.) :
In: Proceedings of the 2022 IEEE Symposium Series on Computational Intelligence (SSCI 2022),
Piscataway: IEEE. pp. 1268-1275
Abstract
We introduce organism networks, which function like a single neural network but are composed of several neural particle networks; while each particle network fulfils the role of a single weight application within the organism network, it is also trained to self-replicate its own weights. As organism networks feature vastly more parameters than simpler architectures, we perform our initial experiments on an arithmetic task as well as on simplified MNIST-dataset classification as a collective. We observe that individual particle networks tend to specialise in either of the tasks and that the ones fully specialised in the secondary task may be dropped from the network without hindering the computational accuracy of the primary task. This leads to the discovery of a novel pruning-strategy for sparse neural networks.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Keywords: | Neural networks ; Collaboration ; Computer architecture ; Organisms ; Task analysis ; Computational intelligence ; Arithmetic |
| Faculties: | Mathematics, Computer Science and Statistics > Computer Science |
| Subjects: | 000 Computer science, information and general works > 004 Data processing computer science |
| ISBN: | 978-1-6654-8768-9 |
| Place of Publication: | Piscataway |
| Item ID: | 124803 |
| Date Deposited: | 04. Nov 2025 12:22 |
| Last Modified: | 04. Nov 2025 12:22 |
