Why it matters:
- More reliable than random-initialized models
- Lower tuning overhead, less trial and error
- Stronger stability across runs
📄 Read more: https://t.co/p6waQBYwTF
2/2
What is the minimally effective way to construct a reservoir computer that can learn chaotic dynamics?
In my new paper, I demonstrate that deterministically constructed reservoirs with minimal topologies (MESNs) consistently outperform conventional random-initialized ESNs.
1/2
Huge thanks to @isp_uv_es for hosting my Julia presentation! It's been an incredible journey as a visiting researcher at this fantastic institution as part of the @ELLISforEurope program. Grateful for the support, knowledge, and connections made during my time here!
Finally cleaned up some code that I had around for ~2 years now, and made it into a #JuliaLang package: introducing CellularAutomata.jl (https://t.co/LmaJOO0Rm6).
I'm happy to share my talk for #JuliaCon 2021! This is a quick introduction to ReservoirComputing.jl, showing how you can build and train and Echo State Network and use it to predict a chaotic system in Julia using no more than 5 lines of code:
https://t.co/hBD7DHQ98F
I'm very excited to share that tomorrow I'll be starting a PhD @UniLeipzig, working with @Rsc4Earth and @Sca_DS to investigate applications of Reservoir Computing to ecosystem anomalies!
Tomorrow I will officially start my #GSoC project with #JuliaLang and #SciML, working on ReservoirComputing.jl. If you are curious about what Reservoir Computing is, then the following post is just for you! https://t.co/viSZJlBAH5