Seq103 unifies NEAT-style neuroevolution for sequence classification, retaining 82-87% of baseline accuracy with 12x to 160k x fewer parameters on text and UCR time-series data.
, author Kallada, M
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Seq103: A Unified Neuroevolution Framework for Compact Sequence Architecture Discovery
Seq103 unifies NEAT-style neuroevolution for sequence classification, retaining 82-87% of baseline accuracy with 12x to 160k x fewer parameters on text and UCR time-series data.