Francesco Martinuzzi

Minimal deterministic echo state networks outperform random reservoirs in learning chaotic dynamics Machine learning (ML) is widely used to model chaotic systems. Among ML approaches, echo state networks (ESNs) have received considerable attention due to their

Why it matters:
- More reliable than random-initialized models
- Lower tuning overhead, less trial and error
- Stronger stability across runs

📄 Read more: pubs.aip.org/aip/cha/arti...

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

So, we managed to survive review and published it is: npg.copernicus.org/articles/31/...

SpectralIndices.jl v0.2.0 is out!
Dive into enriched docs with DataFrames & YAXArrays examples and enjoy significantly faster index computations. Upgrade now for a smoother experience!
#JuliaLang #SpectralIndices

Happy for PhD cand Francesco Martinuzzi for having is paper in discussion comparing various RNNs in modelling vegetation responses ... there are still arguments for #ESNs, but all models struggled to capture impact of extremes. More research needed! 🌱🌍 🔗https://doi.org/10.5194/egusphere-2023-2368