Jason A. Platt
Research Subjects
Jason A. Platt has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Stephen Gregory Penny
Sofar Ocean Technologies, 28 Pier Annex, San Francisco, CA 94105, USA
7 shared publications - Timothy A. Smith
Cooperative Institute for Research in Environmental Sciences (CIRES) at the University of Colorado Boulder Boulder CO USA
5 shared publications - Tse-Chun Chen
Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
5 shared publications - Henry D. I. Abarbanel
Department of Physics and Marine Physical Laboratory, Scripps Institution of Oceanography, University of California, San Diego, Mail Code 0402, La Jolla, California 92093-0402
3 shared publications Adrian Wong
Department of Physics, University of California San Diego, 9500 Gilman Drive, La Jolla, California 92093, USA
2 shared publicationsHenry D. I. Abarbanel
2 shared publications
Latest Publications
- Open Access
- 1 April 2025
- Published by American Meteorological Society in Artificial Intelligence for the Earth Systems
- Vol. 4 (2)
Machine Learned Empirical Numerical Integrator from Simulated Data
Recently, a number of state-of-the-art surrogate machine learning (ML) models have been designed for global weather and climate prediction, which have been trained using reanalysis data products. Reanalysis data products are constructed using numerical model simulations that combine numerical integration of partial differential equations and parameterization schemes. These products are typically only archived and made available using coarsened spatial and temporal resolutions. This study explores the impact of the numerical generation methods used to produce the training datasets and the temporal resolution of those datasets on machine learning surrogate models. Using the nonlinear vector autoregression (NVAR) machine as an explainable ML technique, simple dynamical systems are emulated with ML models trained on data produced by three classical numerical integration schemes. NVAR is validated as a skillful ML method, capable of producing accurate predictions and, more importantly, reconstructing both the underlying dynamics and the numerical integration scheme used to generate the training data. However, the machine fails to generalize predictions on unseen test data generated by different numerical integration schemes, despite the underlying dynamical system being the same. This result provides a word of caution for the growing field of machine learning emulation of weather and climate dynamics. Furthermore, we illustrate using NVAR that training on temporally coarsened data may increase the required complexity of ML models and potentially introduce new numerical challenges. Finally, we discover that empirical integration schemes with arbitrary time-stepping sizes can be constructed directly from the data, which implies a potential for the development of empirical numerical integration schemes.
- 1 April 2025
- Published by American Meteorological Society in Artificial Intelligence for the Earth Systems
- Vol. 4 (2)
Open AccessMachine Learned Empirical Numerical Integrator from Simulated Data
Recently, a number of state-of-the-art surrogate machine learning (ML) models have been designed for global weather and climate prediction, which have been trained using reanalysis data products. Reanalysis data products are constructed using numerical model simulations that combine numerical integration of partial differential equations and parameterization schemes. These products are typically only archived and made available using coarsened spatial and temporal resolutions. This study explores the impact of the numerical generation methods used to produce the training datasets and the temporal resolution of those datasets on machine learning surrogate models. Using the nonlinear vector autoregression (NVAR) machine as an explainable ML technique, simple dynamical systems are emulated with ML models trained on data produced by three classical numerical integration schemes. NVAR is validated as a skillful ML method, capable of producing accurate predictions and, more importantly, reconstructing both the underlying dynamics and the numerical integration scheme used to generate the training data. However, the machine fails to generalize predictions on unseen test data generated by different numerical integration schemes, despite the underlying dynamical system being the same. This result provides a word of caution for the growing field of machine learning emulation of weather and climate dynamics. Furthermore, we illustrate using NVAR that training on temporally coarsened data may increase the required complexity of ML models and potentially introduce new numerical challenges. Finally, we discover that empirical integration schemes with arbitrary time-stepping sizes can be constructed directly from the data, which implies a potential for the development of empirical numerical integration schemes.
- Open Access
- 12 November 2024
- Published by The Open Journal in Journal of Open Source Software
- Vol. 9 (103) , 7286
xesn: Echo state networks powered by Xarray and Dask
Xesn is a Python package that allows scientists to easily design Echo State Networks (ESNs) for forecasting problems. ESNs are a Recurrent Neural Network architecture introduced by Jaeger (2001) that are part of a class of techniques termed Reservoir Computing. One defining characteristic of these techniques is that all internal weights are determined by a handful of global, scalar parameters, thereby avoiding problems during backpropagation and reducing training time significantly. Because this architecture is conceptually simple, many scientists implement ESNs from scratch, leading to questions about computational performance. Xesn offers a straightforward, standard implementation of ESNs that operates efficiently on CPU and GPU hardware. The package leverages optimization tools to automate the parameter selection process, so that scientists can reduce the time finding a good architecture and focus on using ESNs for their domain application. Importantly, the package flexibly handles forecasting tasks for out-of-core, multi-dimensional datasets, eliminating the need to write parallel programming code. Xesn was initially developed to handle the problem of forecasting weather dynamics, and so it integrates naturally with Python packages that have become familiar to weather and climate scientists such as Xarray (Hoyer & Hamman, 2017). However, the software is ultimately general enough to be utilized in other domains where ESNs have been useful, such as in signal processing (Jaeger & Haas, 2004).
- 12 November 2024
- Published by The Open Journal in Journal of Open Source Software
- Vol. 9 (103) , 7286
Open Accessxesn: Echo state networks powered by Xarray and Dask
Xesn is a Python package that allows scientists to easily design Echo State Networks (ESNs) for forecasting problems. ESNs are a Recurrent Neural Network architecture introduced by Jaeger (2001) that are part of a class of techniques termed Reservoir Computing. One defining characteristic of these techniques is that all internal weights are determined by a handful of global, scalar parameters, thereby avoiding problems during backpropagation and reducing training time significantly. Because this architecture is conceptually simple, many scientists implement ESNs from scratch, leading to questions about computational performance. Xesn offers a straightforward, standard implementation of ESNs that operates efficiently on CPU and GPU hardware. The package leverages optimization tools to automate the parameter selection process, so that scientists can reduce the time finding a good architecture and focus on using ESNs for their domain application. Importantly, the package flexibly handles forecasting tasks for out-of-core, multi-dimensional datasets, eliminating the need to write parallel programming code. Xesn was initially developed to handle the problem of forecasting weather dynamics, and so it integrates naturally with Python packages that have become familiar to weather and climate scientists such as Xarray (Hoyer & Hamman, 2017). However, the software is ultimately general enough to be utilized in other domains where ESNs have been useful, such as in signal processing (Jaeger & Haas, 2004).
- Open Access
- 6 December 2023
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 15 (12)
Temporal Subsampling Diminishes Small Spatial Scales in Recurrent Neural Network Emulators of Geophysical Turbulence
The immense computational cost of traditional numerical weather and climate models has sparked the development of machine learning (ML) based emulators. Because ML methods benefit from long records of training data, it is common to use data sets that are temporally subsampled relative to the time steps required for the numerical integration of differential equations. Here, we investigate how this often overlooked processing step affects the quality of an emulator's predictions. We implement two ML architectures from a class of methods called reservoir computing: (a) a form of Nonlinear Vector Autoregression (NVAR), and (b) an Echo State Network (ESN). Despite their simplicity, it is well documented that these architectures excel at predicting low dimensional chaotic dynamics. We are therefore motivated to test these architectures in an idealized setting of predicting high dimensional geophysical turbulence as represented by Surface Quasi‐Geostrophic dynamics. In all cases, subsampling the training data consistently leads to an increased bias at small spatial scales that resembles numerical diffusion. Interestingly, the NVAR architecture becomes unstable when the temporal resolution is increased, indicating that the polynomial based interactions are insufficient at capturing the detailed nonlinearities of the turbulent flow. The ESN architecture is found to be more robust, suggesting a benefit to the more expensive but more general structure. Spectral errors are reduced by including a penalty on the kinetic energy density spectrum during training, although the subsampling related errors persist. Future work is warranted to understand how the temporal resolution of training data affects other ML architectures.
- 6 December 2023
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 15 (12)
Open AccessTemporal Subsampling Diminishes Small Spatial Scales in Recurrent Neural Network Emulators of Geophysical Turbulence
The immense computational cost of traditional numerical weather and climate models has sparked the development of machine learning (ML) based emulators. Because ML methods benefit from long records of training data, it is common to use data sets that are temporally subsampled relative to the time steps required for the numerical integration of differential equations. Here, we investigate how this often overlooked processing step affects the quality of an emulator's predictions. We implement two ML architectures from a class of methods called reservoir computing: (a) a form of Nonlinear Vector Autoregression (NVAR), and (b) an Echo State Network (ESN). Despite their simplicity, it is well documented that these architectures excel at predicting low dimensional chaotic dynamics. We are therefore motivated to test these architectures in an idealized setting of predicting high dimensional geophysical turbulence as represented by Surface Quasi‐Geostrophic dynamics. In all cases, subsampling the training data consistently leads to an increased bias at small spatial scales that resembles numerical diffusion. Interestingly, the NVAR architecture becomes unstable when the temporal resolution is increased, indicating that the polynomial based interactions are insufficient at capturing the detailed nonlinearities of the turbulent flow. The ESN architecture is found to be more robust, suggesting a benefit to the more expensive but more general structure. Spectral errors are reduced by including a penalty on the kinetic energy density spectrum during training, although the subsampling related errors persist. Future work is warranted to understand how the temporal resolution of training data affects other ML architectures.
- 1 October 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (10)
Constraining chaos: Enforcing dynamical invariants in the training of reservoir computers
Drawing on ergodic theory, we introduce a novel training method for machine learning based forecasting methods for chaotic dynamical systems. The training enforces dynamical invariants—such as the Lyapunov exponent spectrum and the fractal dimension—in the systems of interest, enabling longer and more stable forecasts when operating with limited data. The technique is demonstrated in detail using reservoir computing, a specific kind of recurrent neural network. Results are given for the Lorenz 1996 chaotic dynamical system and a spectral quasi-geostrophic model of the atmosphere, both typical test cases for numerical weather prediction.
- 1 October 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (10)
Constraining chaos: Enforcing dynamical invariants in the training of reservoir computers
Drawing on ergodic theory, we introduce a novel training method for machine learning based forecasting methods for chaotic dynamical systems. The training enforces dynamical invariants—such as the Lyapunov exponent spectrum and the fractal dimension—in the systems of interest, enabling longer and more stable forecasts when operating with limited data. The technique is demonstrated in detail using reservoir computing, a specific kind of recurrent neural network. Results are given for the Lorenz 1996 chaotic dynamical system and a spectral quasi-geostrophic model of the atmosphere, both typical test cases for numerical weather prediction.
- Top Cited Papers
- 29 June 2022
- Published by Elsevier in Neural Networks
- Vol. 153, 530-552
A systematic exploration of reservoir computing for forecasting complex spatiotemporal dynamics
- 29 June 2022
- Published by Elsevier in Neural Networks
- Vol. 153, 530-552
Top Cited PapersA systematic exploration of reservoir computing for forecasting complex spatiotemporal dynamics
- 16 June 2022
- Published by MIT Press in Neural Computation
- Vol. 34 (7) , 1545-1587
Reduced-Dimension, Biophysical Neuron Models Constructed From Observed Data
Using methods from nonlinear dynamics and interpolation techniques from applied mathematics, we show how to use data alone to construct discrete time dynamical rules that forecast observed neuron properties. These data may come from simulations of a Hodgkin-Huxley (HH) neuron model or from laboratory current clamp experiments. In each case, the reduced-dimension, data-driven forecasting (DDF) models are shown to predict accurately for times after the training period. When the available observations for neuron preparations are, for example, membrane voltage V(t) only, we use the technique of time delay embedding from nonlinear dynamics to generate an appropriate space in which the full dynamics can be realized. The DDF constructions are reduced-dimension models relative to HH models as they are built on and forecast only observables such as V(t). They do not require detailed specification of ion channels, their gating variables, and the many parameters that accompany an HH model for laboratory measurements, yet all of this important information is encoded in the DDF model. As the DDF models use and forecast only voltage data, they can be used in building networks with biophysical connections. Both gap junction connections and ligand gated synaptic connections among neurons involve presynaptic voltages and induce postsynaptic voltage response. Biophysically based DDF neuron models can replace other reduced-dimension neuron models, say, of the integrate-and-fire type, in developing and analyzing large networks of neurons. When one does have detailed HH model neurons for network components, a reduced-dimension DDF realization of the HH voltage dynamics may be used in network computations to achieve computational efficiency and the exploration of larger biological networks.
- 16 June 2022
- Published by MIT Press in Neural Computation
- Vol. 34 (7) , 1545-1587
Reduced-Dimension, Biophysical Neuron Models Constructed From Observed Data
Using methods from nonlinear dynamics and interpolation techniques from applied mathematics, we show how to use data alone to construct discrete time dynamical rules that forecast observed neuron properties. These data may come from simulations of a Hodgkin-Huxley (HH) neuron model or from laboratory current clamp experiments. In each case, the reduced-dimension, data-driven forecasting (DDF) models are shown to predict accurately for times after the training period. When the available observations for neuron preparations are, for example, membrane voltage V(t) only, we use the technique of time delay embedding from nonlinear dynamics to generate an appropriate space in which the full dynamics can be realized. The DDF constructions are reduced-dimension models relative to HH models as they are built on and forecast only observables such as V(t). They do not require detailed specification of ion channels, their gating variables, and the many parameters that accompany an HH model for laboratory measurements, yet all of this important information is encoded in the DDF model. As the DDF models use and forecast only voltage data, they can be used in building networks with biophysical connections. Both gap junction connections and ligand gated synaptic connections among neurons involve presynaptic voltages and induce postsynaptic voltage response. Biophysically based DDF neuron models can replace other reduced-dimension neuron models, say, of the integrate-and-fire type, in developing and analyzing large networks of neurons. When one does have detailed HH model neurons for network components, a reduced-dimension DDF realization of the HH voltage dynamics may be used in network computations to achieve computational efficiency and the exploration of larger biological networks.
- Open Access
- 4 March 2022
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 14 (3)
Integrating Recurrent Neural Networks With Data Assimilation for Scalable Data‐Driven State Estimation
Data assimilation (DA) is integrated with machine learning in order to perform entirely data‐driven online state estimation. To achieve this, recurrent neural networks (RNNs) are implemented as pretrained surrogate models to replace key components of the DA cycle in numerical weather prediction (NWP), including the conventional numerical forecast model, the forecast error covariance matrix, and the tangent linear and adjoint models. It is shown how these RNNs can be initialized using DA methods to directly update the hidden/reservoir state with observations of the target system. The results indicate that these techniques can be applied to estimate the state of a system for the repeated initialization of short‐term forecasts, even in the absence of a traditional numerical forecast model. Further, it is demonstrated how these integrated RNN‐DA methods can scale to higher dimensions by applying domain localization and parallelization, providing a path for practical applications in NWP.
- 4 March 2022
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 14 (3)
Open AccessIntegrating Recurrent Neural Networks With Data Assimilation for Scalable Data‐Driven State Estimation
Data assimilation (DA) is integrated with machine learning in order to perform entirely data‐driven online state estimation. To achieve this, recurrent neural networks (RNNs) are implemented as pretrained surrogate models to replace key components of the DA cycle in numerical weather prediction (NWP), including the conventional numerical forecast model, the forecast error covariance matrix, and the tangent linear and adjoint models. It is shown how these RNNs can be initialized using DA methods to directly update the hidden/reservoir state with observations of the target system. The results indicate that these techniques can be applied to estimate the state of a system for the repeated initialization of short‐term forecasts, even in the absence of a traditional numerical forecast model. Further, it is demonstrated how these integrated RNN‐DA methods can scale to higher dimensions by applying domain localization and parallelization, providing a path for practical applications in NWP.
- 1 December 2021
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 31 (12) , 123118
Robust forecasting using predictive generalized synchronization in reservoir computing
Reservoir computers (RCs) are a class of recurrent neural networks (RNNs) that can be used for forecasting the future of observed time series data. As with all RNNs, selecting the hyperparameters in the network to yield excellent forecasting presents a challenge when training on new inputs. We analyze a method based on predictive generalized synchronization (PGS) that gives direction in designing and evaluating the architecture and hyperparameters of an RC. To determine the occurrences of PGS, we rely on the auxiliary method to provide a computationally efficient pre-training test that guides hyperparameter selection. We provide a metric for evaluating the RC using the reproduction of the input system’s Lyapunov exponents that demonstrates robustness in prediction.
- 1 December 2021
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 31 (12) , 123118
Robust forecasting using predictive generalized synchronization in reservoir computing
Reservoir computers (RCs) are a class of recurrent neural networks (RNNs) that can be used for forecasting the future of observed time series data. As with all RNNs, selecting the hyperparameters in the network to yield excellent forecasting presents a challenge when training on new inputs. We analyze a method based on predictive generalized synchronization (PGS) that gives direction in designing and evaluating the architecture and hyperparameters of an RC. To determine the occurrences of PGS, we rely on the auxiliary method to provide a computationally efficient pre-training test that guides hyperparameter selection. We provide a metric for evaluating the RC using the reproduction of the input system’s Lyapunov exponents that demonstrates robustness in prediction.
- Open Access
Forecasting Using Reservoir Computing: The Role of Generalized Synchronization
Reservoir computers (RC) are a form of recurrent neural network (RNN) used for forecasting time series data. As with all RNNs, selecting the hyperparameters presents a challenge when training on new inputs. We present a method based on generalized synchronization (GS) that gives direction in designing and evaluating the architecture and hyperparameters of a RC. The 'auxiliary method' for detecting GS provides a pre-training test that guides hyperparameter selection. Furthermore, we provide a metric for a "well trained" RC using the reproduction of the input system's Lyapunov exponents.
Open AccessForecasting Using Reservoir Computing: The Role of Generalized Synchronization
Reservoir computers (RC) are a form of recurrent neural network (RNN) used for forecasting time series data. As with all RNNs, selecting the hyperparameters presents a challenge when training on new inputs. We present a method based on generalized synchronization (GS) that gives direction in designing and evaluating the architecture and hyperparameters of a RC. The 'auxiliary method' for detecting GS provides a pre-training test that guides hyperparameter selection. Furthermore, we provide a metric for a "well trained" RC using the reproduction of the input system's Lyapunov exponents.
- Open Access
Machine Learning Classification Informed by a Functional Biophysical System
We present a novel machine learning architecture for classification suggested by experiments on olfactory systems. The network separates input stimuli, represented as spatially distinct currents, via winnerless competition---a process based on the intrinsic sequential dynamics of the neural system---then uses a support vector machine (SVM) to provide precision to the space-time separation of the output. The combined network uses biophysical models of neurons and shows high discrimination among inputs and robustness to noise. While using the SVM alone does not permit determination of the components of mixtures of classified inputs, the combined network is able to tell the precise concentrations of the constituent parts.
Open AccessMachine Learning Classification Informed by a Functional Biophysical System
We present a novel machine learning architecture for classification suggested by experiments on olfactory systems. The network separates input stimuli, represented as spatially distinct currents, via winnerless competition---a process based on the intrinsic sequential dynamics of the neural system---then uses a support vector machine (SVM) to provide precision to the space-time separation of the output. The combined network uses biophysical models of neurons and shows high discrimination among inputs and robustness to noise. While using the SVM alone does not permit determination of the components of mixtures of classified inputs, the combined network is able to tell the precise concentrations of the constituent parts.