Miguel C. Soriano
@miguelcsoriano.bsky.social
Researcher on Complex Systems, Nonlinear Photonics and Neuro-inspired Information Processing
Reposted by Miguel C. Soriano
📢 La #NitDeLaRecerca ens va deixar un vespre ple de ciència! @miguelcsoriano.bsky.social va obrir amb un pregó sobre IA 🖥 i @rherzoga.bsky.social ens va parlar de medicaments psicodèlics 💊.
Mentrestant, els més petits van gaudir d’experiments sobre complexitat 🔬✨.
Mentrestant, els més petits van gaudir d’experiments sobre complexitat 🔬✨.
September 29, 2025 at 7:09 AM
📢 La #NitDeLaRecerca ens va deixar un vespre ple de ciència! @miguelcsoriano.bsky.social va obrir amb un pregó sobre IA 🖥 i @rherzoga.bsky.social ens va parlar de medicaments psicodèlics 💊.
Mentrestant, els més petits van gaudir d’experiments sobre complexitat 🔬✨.
Mentrestant, els més petits van gaudir d’experiments sobre complexitat 🔬✨.
Físicos buscan la tecnología del futuro: ordenadores de fotones, más rápidos y menos contaminantes www.eldiario.es/1_c07e83?utm...
Físicos buscan la tecnología del futuro: ordenadores de fotones, más rápidos y menos contaminantes
Un equipo internacional de investigadores, liderados desde España, estudia utilizar la luz para sustituir los sistemas digitales en el procesamiento de datos, lo que reduciría la huella de carbono
www.eldiario.es
September 20, 2025 at 1:16 PM
Físicos buscan la tecnología del futuro: ordenadores de fotones, más rápidos y menos contaminantes www.eldiario.es/1_c07e83?utm...
Reposted by Miguel C. Soriano
🚨New preprint !!
Interpreting the training process of a neural network as a temporal network trajectory, we found a regime where such trajectory becomes chaotic. Rather than a nuisance, such chaotic mixing boosts training!
w/ P. Jiménez, @miguelcsoriano.bsky.social
arxiv.org/abs/2506.08523
Interpreting the training process of a neural network as a temporal network trajectory, we found a regime where such trajectory becomes chaotic. Rather than a nuisance, such chaotic mixing boosts training!
w/ P. Jiménez, @miguelcsoriano.bsky.social
arxiv.org/abs/2506.08523
Leveraging chaos in the training of artificial neural networks
Traditional algorithms to optimize artificial neural networks when confronted with a supervised learning task are usually exploitation-type relaxational dynamics such as gradient descent (GD). Here, w...
arxiv.org
June 15, 2025 at 7:16 PM
🚨New preprint !!
Interpreting the training process of a neural network as a temporal network trajectory, we found a regime where such trajectory becomes chaotic. Rather than a nuisance, such chaotic mixing boosts training!
w/ P. Jiménez, @miguelcsoriano.bsky.social
arxiv.org/abs/2506.08523
Interpreting the training process of a neural network as a temporal network trajectory, we found a regime where such trajectory becomes chaotic. Rather than a nuisance, such chaotic mixing boosts training!
w/ P. Jiménez, @miguelcsoriano.bsky.social
arxiv.org/abs/2506.08523