arXiv research

New deep ESN architectures improve memory capacity and prediction accuracy.

Echo State Networks (ESNs) are recurrent neural networks that only train their output layer, thereby precluding the need to backpropagate gradients through time, which leads to significant computational gains. Nevertheless, a common issue in ESNs is determining its hyperparameters, which are crucial in instantiating a …

ESNs trained with Tikhonov least squares approximate ergodic dynamical systems in L2(μ) norm.

Efficient cross-validation method for Echo State Networks reduces validation time complexity.

New method combines long-memory reservoirs for accurate dengue forecasting from short data.

Efficient CV for ESNs improves time series predictions.

Deep models predict intraday electricity prices accurately.

Feedback improves ESN performance by 30-60% across various tasks.

Spatially aware ESN detects anomalies in chaotic time series.

We propose a physics-informed Echo State Network (ESN) to predict the evolution of chaotic systems. Compared to conventional ESNs, the physics-informed ESNs are trained to solve supervised learning tasks while ensuring that their predictions do not violate physical laws. This is achieved by introducing an additional lo…

New analysis reveals optimal regularization for ESNs, avoiding double descent.

Echo State Networks (ESN) are a class of Recurrent Neural Networks (RNN) that has gained substantial popularity due to their effectiveness, ease of use and potential for compact hardware implementation. An ESN contains the three network layers input, reservoir and readout where the reservoir is the truly recurrent netw…

Optical ESNs enable flexible, efficient machine learning with reduced energy.

New framework analyzes temporal features in state space models.

Echo state networks are powerful recurrent neural networks. However, they are often unstable and shaky, making the process of finding an good ESN for a specific dataset quite hard. Obtaining a superb accuracy by using the Echo State Network is a challenging task. We create, develop and implement a family of predictably…

ESNs with transfer learning predict long-term chaotic patterns in spatiotemporal dynamical systems.

ESN model helps understand climate event impacts.

We propose an experimental comparison between Deep Echo State Networks (DeepESNs) and gated Recurrent Neural Networks (RNNs) on multivariate time-series prediction tasks. In particular, we compare reservoir and fully-trained RNNs able to represent signals featured by multiple time-scales dynamics. The analysis is perfo…

Machine learning predicts dam-break flood wave behavior accurately.

Echo state network (ESN) is viewed as a temporal non-orthogonal expansion with pseudo-random parameters. Such expansions naturally give rise to regressors of various relevance to a teacher output. We illustrate that often only a certain amount of the generated echo-regressors effectively explain the variance of the tea…

We analyze generalization in deep learning models using random matrix theory.

Recurrent neural networks (RNNs) have drawn interest from machine learning researchers because of their effectiveness at preserving past inputs for time-varying data processing tasks. To understand the success and limitations of RNNs, it is critical that we advance our analysis of their fundamental memory properties. W…

Long-lead forecasting for spatio-temporal systems can often entail complex nonlinear dynamics that are difficult to specify it a priori. Current statistical methodologies for modeling these processes are often highly parameterized and thus, challenging to implement from a computational perspective. One potential parsim…

Reservoir Computing (RC) refers to a Recurrent Neural Networks (RNNs) framework, frequently used for sequence learning and time series prediction. The RC system consists of a random fixed-weight RNN (the input-hidden reservoir layer) and a classifier (the hidden-output readout layer). Here we focus on the sequence lear…

ESN models predict intraday stock returns efficiently.

Deep learning models predict chaotic Lorenz 96 system accurately.

SORSCNs improve nonstationary data modeling by self-organizing and adjusting network parameters.

HFNO enhances interpretability of turbulent flows through parallel wavenumber bin processing.

Study shows structured reservoirs improve deep ESN performance.

Spatio-temporal data and processes are prevalent across a wide variety of scientific disciplines. These processes are often characterized by nonlinear time dynamics that include interactions across multiple scales of spatial and temporal variability. The data sets associated with many of these processes are increasing …

CogScale benchmarks AI architectures for sequential processing.

Previous studies indicate that nonlinear properties of Gaussian time series with long-range correlations, $u_i$, can be detected and quantified by studying the correlations in the magnitude series $|u_i|$, i.e., the ``volatility''. However, the origin for this empirical observation still remains unclear, and the exact …

Research tackles unequal length time series for classification.

Lie-Butcher (LB) series are formal power series expressed in terms of trees and forests. On the geometric side LB-series generalizes classical B-series from Euclidean spaces to Lie groups and homogeneous manifolds. On the algebraic side, B-series are based on pre-Lie algebras and the Butcher-Connes-Kreimer Hopf algebra…

The explosion of time series data in recent years has brought a flourish of new time series analysis methods, for forecasting, clustering, classification and other tasks. The evaluation of these new methods requires either collecting or simulating a diverse set of time series benchmarking data to enable reliable compar…

Study series invariants of plumbed 3-manifolds using root lattices.

New formula and properties of inverted Habiro series derived from GM series.

Paper introduces novel distances for clustering ordinal time series.

MDF represents time series motifs as images for improved classification.

MPPN network improves long-term time series forecasting accuracy.

Overview of high-dimensional time series regression methods.

Archive of 20 time series datasets for forecasting evaluation.

Paper proposes a robust time series classification method using ResNet and Recurrence Plots.

New method uses Transformers for flu forecasting.

Improved prediction of hierarchical time series using structured regularization.

New kernel handles irregularly-spaced multivariate time series.

Capturing the dynamical properties of time series concisely as interpretable feature vectors can enable efficient clustering and classification for time-series applications across science and industry. Selecting an appropriate feature-based representation of time series for a given application can be achieved through s…

Modeling regime shifts in co-evolving time series with interactions and time-dependency.