home.social

#reservoircomputing — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #reservoircomputing, aggregated by home.social.

  1. Representation learning often emphasizes metric preservation. We instead build Symplectic structural invariance directly into the representation.

    arxiv.org/abs/2512.19409

    We embed Hamiltonian/symplectic geometry by making the RNN state dynamics a symplectomorphism, which preserves Legendre duality (information geometry) through time. This yields structure-preserving representations enforced by the latent dynamics, rather than imposed indirectly via the output.

    #ReservoirComputing #RepresentationLearning #InformationGeometry #SymplecticGeometry #HamiltonianDynamics #GeometricDeepLearning #DynamicalSystems #PhysicsInformedML

  2. For #ReservoirComputing lovers,I found a challenging #attractor, the Thomas':
    dx = np.sin(y) - b * x
    dy = np.sin(z) - b * y
    dz = np.sin(x) - b * z
    x += dx * dt
    y += dy * dt
    z += dz * dt
    3k neurons yet 67% correlation. Here is the code for you github.com/alecrimi/magic_rese

  3. For lovers,I found a challenging , the Thomas':
    dx = np.sin(y) - b * x
    dy = np.sin(z) - b * y
    dz = np.sin(x) - b * z
    x += dx * dt
    y += dy * dt
    z += dz * dt
    3k neurons yet 67% correlation. Here is the code for you github.com/alecrimi/magic_rese

  4. For #ReservoirComputing lovers,I found a challenging #attractor, the Thomas':
    dx = np.sin(y) - b * x
    dy = np.sin(z) - b * y
    dz = np.sin(x) - b * z
    x += dx * dt
    y += dy * dt
    z += dz * dt
    3k neurons yet 67% correlation. Here is the code for you github.com/alecrimi/magic_rese

  5. For #ReservoirComputing lovers,I found a challenging #attractor, the Thomas':
    dx = np.sin(y) - b * x
    dy = np.sin(z) - b * y
    dz = np.sin(x) - b * z
    x += dx * dt
    y += dy * dt
    z += dz * dt
    3k neurons yet 67% correlation. Here is the code for you github.com/alecrimi/magic_rese

  6. For #ReservoirComputing lovers,I found a challenging #attractor, the Thomas':
    dx = np.sin(y) - b * x
    dy = np.sin(z) - b * y
    dz = np.sin(x) - b * z
    x += dx * dt
    y += dy * dt
    z += dz * dt
    3k neurons yet 67% correlation. Here is the code for you github.com/alecrimi/magic_rese

  7. Designing nonlinearity in a current-starved ring oscillator for reservoir computing hardware

    Jaeger, H. & Haas, H. Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication. Scienc…
    #NewsBeep #News #Computing #AU #Australia #Current-starvedringoscillator #Energyscienceandtechnology #Engineering #Frequency #HumanitiesandSocialSciences #learning #multidisciplinary #Nonlinear #Physics #Reservoircomputing #Science #Technology
    newsbeep.com/au/182302/

  8. Designing nonlinearity in a current-starved ring oscillator for reservoir computing hardware

    Jaeger, H. & Haas, H. Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication. Sc…
    #NewsBeep #News #Computing #Current-starvedringoscillator #Energyscienceandtechnology #Engineering #Frequency #HumanitiesandSocialSciences #learning #multidisciplinary #Nonlinear #Physics #Reservoircomputing #Science #Technology #UK #UnitedKingdom
    newsbeep.com/uk/172553/

  9. Li #attractor=chaotic vibes like Lorenz/Chen use its wild dynamics for #ReservoirComputing to process time series.
    The truth: just use it because it looks cool⚔🏴‍☠️🌊
    def dequan_li(x,y,z):
    dx = a*(y - x) + y*z
    dy = b*x - x*z+y
    dz = c*z + x*y/3
    return dx,dy,dz

  10. Li =chaotic vibes like Lorenz/Chen use its wild dynamics for to process time series.
    The truth: just use it because it looks cool⚔🏴‍☠️🌊
    def dequan_li(x,y,z):
    dx = a*(y - x) + y*z
    dy = b*x - x*z+y
    dz = c*z + x*y/3
    return dx,dy,dz

  11. Li #attractor=chaotic vibes like Lorenz/Chen use its wild dynamics for #ReservoirComputing to process time series.
    The truth: just use it because it looks cool⚔🏴‍☠️🌊
    def dequan_li(x,y,z):
    dx = a*(y - x) + y*z
    dy = b*x - x*z+y
    dz = c*z + x*y/3
    return dx,dy,dz

  12. Li #attractor=chaotic vibes like Lorenz/Chen use its wild dynamics for #ReservoirComputing to process time series.
    The truth: just use it because it looks cool⚔🏴‍☠️🌊
    def dequan_li(x,y,z):
    dx = a*(y - x) + y*z
    dy = b*x - x*z+y
    dz = c*z + x*y/3
    return dx,dy,dz

  13. Li #attractor=chaotic vibes like Lorenz/Chen use its wild dynamics for #ReservoirComputing to process time series.
    The truth: just use it because it looks cool⚔🏴‍☠️🌊
    def dequan_li(x,y,z):
    dx = a*(y - x) + y*z
    dy = b*x - x*z+y
    dz = c*z + x*y/3
    return dx,dy,dz

  14. Summer ☀️ read on Computo: a new publication on reservoir computing in R!

    Reservoir computing is a machine learning approach that relies on mapping inputs to higher dimensional spaces through a non-linear dynamical system (the reservoir), for example using a deep recurrent neural network and training only its final layer.

    In this new publication, Thomas Ferté and co-authors Kalidou Ba, Dan Dutartre, Pierrick Legrand, Vianney Jouhet, Rodolphe Thiébaut, Xavier Hinaut and Boris P. Hejblum present the reservoirnet package, which is the first implementation of reservoir computing in R (rather then the existing Python and Julia).

    The article also serves as an introduction to reservoir computing and illustrates its usefulness as well as the usage of the package on several real-case applications, including forecasting the number of COVID-19 hospitalizations at Bordeaux University Hospital using public data (epidemiological statistics, weather data) and hospital-level data (hospitalizations, ICU admission, ambulance service and ER notes). On this example, the authors also show how the weights of the connection between the input and the output layers can be used to compute feature importances.

    The paper and accompanying R code are available at doi.org/10.57750/arxn-6z34

    reservoirnet is available at cran.r-project.org/package=res

    #machineLearning #reservoirComputing #Rstats #openScience #openSource #openAccess

  15. Summer ☀️ read on Computo: a new publication on reservoir computing in R!

    Reservoir computing is a machine learning approach that relies on mapping inputs to higher dimensional spaces through a non-linear dynamical system (the reservoir), for example using a deep recurrent neural network and training only its final layer.

    In this new publication, Thomas Ferté and co-authors Kalidou Ba, Dan Dutartre, Pierrick Legrand, Vianney Jouhet, Rodolphe Thiébaut, Xavier Hinaut and Boris P. Hejblum present the reservoirnet package, which is the first implementation of reservoir computing in R (rather then the existing Python and Julia).

    The article also serves as an introduction to reservoir computing and illustrates its usefulness as well as the usage of the package on several real-case applications, including forecasting the number of COVID-19 hospitalizations at Bordeaux University Hospital using public data (epidemiological statistics, weather data) and hospital-level data (hospitalizations, ICU admission, ambulance service and ER notes). On this example, the authors also show how the weights of the connection between the input and the output layers can be used to compute feature importances.

    The paper and accompanying R code are available at doi.org/10.57750/arxn-6z34

    reservoirnet is available at cran.r-project.org/package=res

    #machineLearning #reservoirComputing #Rstats #openScience #openSource #openAccess

  16. Summer ☀️ read on Computo: a new publication on reservoir computing in R!

    Reservoir computing is a machine learning approach that relies on mapping inputs to higher dimensional spaces through a non-linear dynamical system (the reservoir), for example using a deep recurrent neural network and training only its final layer.

    In this new publication, Thomas Ferté and co-authors Kalidou Ba, Dan Dutartre, Pierrick Legrand, Vianney Jouhet, Rodolphe Thiébaut, Xavier Hinaut and Boris P. Hejblum present the reservoirnet package, which is the first implementation of reservoir computing in R (rather then the existing Python and Julia).

    The article also serves as an introduction to reservoir computing and illustrates its usefulness as well as the usage of the package on several real-case applications, including forecasting the number of COVID-19 hospitalizations at Bordeaux University Hospital using public data (epidemiological statistics, weather data) and hospital-level data (hospitalizations, ICU admission, ambulance service and ER notes). On this example, the authors also show how the weights of the connection between the input and the output layers can be used to compute feature importances.

    The paper and accompanying R code are available at doi.org/10.57750/arxn-6z34

    reservoirnet is available at cran.r-project.org/package=res

    #machineLearning #reservoirComputing #Rstats #openScience #openSource #openAccess

  17. Summer ☀️ read on Computo: a new publication on reservoir computing in R!

    Reservoir computing is a machine learning approach that relies on mapping inputs to higher dimensional spaces through a non-linear dynamical system (the reservoir), for example using a deep recurrent neural network and training only its final layer.

    In this new publication, Thomas Ferté and co-authors Kalidou Ba, Dan Dutartre, Pierrick Legrand, Vianney Jouhet, Rodolphe Thiébaut, Xavier Hinaut and Boris P. Hejblum present the reservoirnet package, which is the first implementation of reservoir computing in R (rather then the existing Python and Julia).

    The article also serves as an introduction to reservoir computing and illustrates its usefulness as well as the usage of the package on several real-case applications, including forecasting the number of COVID-19 hospitalizations at Bordeaux University Hospital using public data (epidemiological statistics, weather data) and hospital-level data (hospitalizations, ICU admission, ambulance service and ER notes). On this example, the authors also show how the weights of the connection between the input and the output layers can be used to compute feature importances.

    The paper and accompanying R code are available at doi.org/10.57750/arxn-6z34

    reservoirnet is available at cran.r-project.org/package=res

    #machineLearning #reservoirComputing #Rstats #openScience #openSource #openAccess

  18. Summer ☀️ read on Computo: a new publication on reservoir computing in R!

    Reservoir computing is a machine learning approach that relies on mapping inputs to higher dimensional spaces through a non-linear dynamical system (the reservoir), for example using a deep recurrent neural network and training only its final layer.

    In this new publication, Thomas Ferté and co-authors Kalidou Ba, Dan Dutartre, Pierrick Legrand, Vianney Jouhet, Rodolphe Thiébaut, Xavier Hinaut and Boris P. Hejblum present the reservoirnet package, which is the first implementation of reservoir computing in R (rather then the existing Python and Julia).

    The article also serves as an introduction to reservoir computing and illustrates its usefulness as well as the usage of the package on several real-case applications, including forecasting the number of COVID-19 hospitalizations at Bordeaux University Hospital using public data (epidemiological statistics, weather data) and hospital-level data (hospitalizations, ICU admission, ambulance service and ER notes). On this example, the authors also show how the weights of the connection between the input and the output layers can be used to compute feature importances.

    The paper and accompanying R code are available at doi.org/10.57750/arxn-6z34

    reservoirnet is available at cran.r-project.org/package=res

    #machineLearning #reservoirComputing #Rstats #openScience #openSource #openAccess

  19. Die #DeutschePhysikalischeGesellschaft hat auf ihrer Frühjahrstagung der Sektion #KondensierteMaterie drei Physiker mit Forschungspreisen für ihre wissenschaftlichen Arbeiten an der #UniMainz ausgezeichnet.

    🎉 Herzlichen Glückwunsch an Dr. Libor Šmejkal zum Walter-Schottky-Preis 2025, an Dr. Robin R. Neumann zum INNOMAG Dissertationspreis 2025 und Grischa Beneke zum INNOMAG Master-Preis 2025 👉 phmi.uni-mainz.de/auszeichnung

    #Physik #Altermagnetismus #Magnonen #ReservoirComputing

  20. Die #DeutschePhysikalischeGesellschaft hat auf ihrer Frühjahrstagung der Sektion #KondensierteMaterie drei Physiker mit Forschungspreisen für ihre wissenschaftlichen Arbeiten an der #UniMainz ausgezeichnet.

    🎉 Herzlichen Glückwunsch an Dr. Libor Šmejkal zum Walter-Schottky-Preis 2025, an Dr. Robin R. Neumann zum INNOMAG Dissertationspreis 2025 und Grischa Beneke zum INNOMAG Master-Preis 2025 👉 phmi.uni-mainz.de/auszeichnung

    #Physik #Altermagnetismus #Magnonen #ReservoirComputing

  21. Die #DeutschePhysikalischeGesellschaft hat auf ihrer Frühjahrstagung der Sektion #KondensierteMaterie drei Physiker mit Forschungspreisen für ihre wissenschaftlichen Arbeiten an der #UniMainz ausgezeichnet.

    🎉 Herzlichen Glückwunsch an Dr. Libor Šmejkal zum Walter-Schottky-Preis 2025, an Dr. Robin R. Neumann zum INNOMAG Dissertationspreis 2025 und Grischa Beneke zum INNOMAG Master-Preis 2025 👉 phmi.uni-mainz.de/auszeichnung

    #Physik #Altermagnetismus #Magnonen #ReservoirComputing

  22. Die #DeutschePhysikalischeGesellschaft hat auf ihrer Frühjahrstagung der Sektion #KondensierteMaterie drei Physiker mit Forschungspreisen für ihre wissenschaftlichen Arbeiten an der #UniMainz ausgezeichnet.

    🎉 Herzlichen Glückwunsch an Dr. Libor Šmejkal zum Walter-Schottky-Preis 2025, an Dr. Robin R. Neumann zum INNOMAG Dissertationspreis 2025 und Grischa Beneke zum INNOMAG Master-Preis 2025 👉 phmi.uni-mainz.de/auszeichnung

    #Physik #Altermagnetismus #Magnonen #ReservoirComputing

  23. Die #DeutschePhysikalischeGesellschaft hat auf ihrer Frühjahrstagung der Sektion #KondensierteMaterie drei Physiker mit Forschungspreisen für ihre wissenschaftlichen Arbeiten an der #UniMainz ausgezeichnet.

    🎉 Herzlichen Glückwunsch an Dr. Libor Šmejkal zum Walter-Schottky-Preis 2025, an Dr. Robin R. Neumann zum INNOMAG Dissertationspreis 2025 und Grischa Beneke zum INNOMAG Master-Preis 2025 👉 phmi.uni-mainz.de/auszeichnung

    #Physik #Altermagnetismus #Magnonen #ReservoirComputing

  24. Updates on the 🧠 effective connectivity library.
    1. Added causality as nonlinear :
    github.com/alecrimi/effconnpy/
    2. Now lables on to the nodes
    3. Added directionality shown with colors from blue to white (white as the arrows)
    github.com/alecrimi/effconnpy/

  25. Updates on the 🧠#brain effective connectivity library.
    1. Added #Reservoircomputing causality as nonlinear #Granger:
    github.com/alecrimi/effconnpy/
    2. Now lables on to the nodes
    3. Added directionality shown with colors from blue to white (white as the arrows)
    github.com/alecrimi/effconnpy/

  26. Updates on the 🧠#brain effective connectivity library.
    1. Added #Reservoircomputing causality as nonlinear #Granger:
    github.com/alecrimi/effconnpy/
    2. Now lables on to the nodes
    3. Added directionality shown with colors from blue to white (white as the arrows)
    github.com/alecrimi/effconnpy/

  27. Updates on the 🧠#brain effective connectivity library.
    1. Added #Reservoircomputing causality as nonlinear #Granger:
    github.com/alecrimi/effconnpy/
    2. Now lables on to the nodes
    3. Added directionality shown with colors from blue to white (white as the arrows)
    github.com/alecrimi/effconnpy/

  28. Updates on the 🧠#brain effective connectivity library.
    1. Added #Reservoircomputing causality as nonlinear #Granger:
    github.com/alecrimi/effconnpy/
    2. Now lables on to the nodes
    3. Added directionality shown with colors from blue to white (white as the arrows)
    github.com/alecrimi/effconnpy/

  29. #simplicialcomplex + #Causality +#Reservoircomputing:
    "Higher-order Granger reservoir computing: simultaneously achieving scalable complex structures inference and accurate dynamics prediction" nature.com/articles/s41467-024

    #dynamicalsystem #ML #AI

  30. + +:
    "Higher-order Granger reservoir computing: simultaneously achieving scalable complex structures inference and accurate dynamics prediction" nature.com/articles/s41467-024

  31. #simplicialcomplex + #Causality +#Reservoircomputing:
    "Higher-order Granger reservoir computing: simultaneously achieving scalable complex structures inference and accurate dynamics prediction" nature.com/articles/s41467-024

    #dynamicalsystem #ML #AI

  32. #simplicialcomplex + #Causality +#Reservoircomputing:
    "Higher-order Granger reservoir computing: simultaneously achieving scalable complex structures inference and accurate dynamics prediction" nature.com/articles/s41467-024

    #dynamicalsystem #ML #AI

  33. #simplicialcomplex + #Causality +#Reservoircomputing:
    "Higher-order Granger reservoir computing: simultaneously achieving scalable complex structures inference and accurate dynamics prediction" nature.com/articles/s41467-024

    #dynamicalsystem #ML #AI

  34. High-Speed Reservoir Computing With Integrated Laser Graded Artificial Neurons - So-called neuromorphic computing involves the use of physical artificial neurons t... - hackaday.com/2025/03/13/high-s #reservoircomputing #neuromorphic #science

  35. High-Speed Reservoir Computing With Integrated Laser Graded Artificial Neurons - So-called neuromorphic computing involves the use of physical artificial neurons t... - hackaday.com/2025/03/13/high-s #reservoircomputing #neuromorphic #science

  36. High-Speed Reservoir Computing With Integrated Laser Graded Artificial Neurons - So-called neuromorphic computing involves the use of physical artificial neurons t... - hackaday.com/2025/03/13/high-s #reservoircomputing #neuromorphic #science

  37. High-Speed Reservoir Computing With Integrated Laser Graded Artificial Neurons - So-called neuromorphic computing involves the use of physical artificial neurons t... - hackaday.com/2025/03/13/high-s #reservoircomputing #neuromorphic #science

  38. High-Speed Reservoir Computing With Integrated Laser Graded Artificial Neurons - So-called neuromorphic computing involves the use of physical artificial neurons t... - hackaday.com/2025/03/13/high-s #reservoircomputing #neuromorphic #science

  39. Energy-saving computing with magnetic whirls: Brownian #ReservoirComputing allows to detect human hand gestures on the basis of diffusion and displacement of #skyrmions // #physics #spintronics #magnetism #computing #TopDyn
    @NatureComms @uni_mainz_eng
    nachrichten.idw-online.de/2024