Coincidex
Transfer-aware, replay-free continual learning for PyTorch
Listed by
Rakib I.80
About this launch
Most continual-learning methods either forget (fine-tuning) or isolate and forfeit transfer (fixed masks, adapters). Coincidex does both: similar tasks share sub-networks for forward transfer, dissimilar ones stay isolated so old skills survive — with no replay buffer. A tiny, drop-in PyTorch layer (torch + numpy, CPU-friendly) that learns task similarity from context. It's also honest: reproducible benchmarks and a clear map of where it works and where it doesn't. Apache-2.0 license.
- Launch date
- Mon, Jul 20, 2026
- Founder trust
- 80 · Trusted
- Category
- Developer Tools
- Website
- github.com
What is LaunchPact?
LaunchPact is a network of founders who show up for each other on Product Hunt launch day. Back launches you believe in — we'll send a reminder when it's time to upvote. Free to join.
Support on launch day
Like what you see? Join LaunchPact free to back Coincidex. We'll remind you on Product Hunt launch day — takes about 30 seconds.
3 expected upvotes
0 backings · 3 pacts
Already on LaunchPact? Sign in
Launch momentum
Screenshot-verified Product Hunt upvotes
