Reservoir computing systems, a class of recurrent neural networks, have recently been exploited for model-free, data-based prediction of the state evolution of a variety of chaotic dynamical systems. The prediction horizon demonstrated has been about half dozen Lyapunov time. Is it possible to significantly extend the prediction time beyond what has been achieved so far? We articulate a scheme incorporating time-dependent but sparse data inputs into reservoir computing and demonstrate that such rare "updates" of the actual state practically enable an arbitrarily long prediction horizon for a variety of chaotic systems. A physical understanding based on the theory of temporal synchronization is developed.
University of Cambridge（剑桥大学）；The Alan Turing Institute（阿兰·图灵研究所）；Technical University of Munich（慕尼黑工业大学）
An approach to the time-accurate prediction of chaotic solutions is by learning temporal patterns from data. Echo State Networks (ESNs), which are a class of Reservoir Computing, can accurately predict the chaotic dynamics well beyond the predictability time. Existing studies, however, also showed that small changes in the hyperparameters may markedly affect the network's performance. The aim of this paper is to assess and improve the robustness of Echo State Networks for the time-accurate prediction of chaotic solutions. The goal is three-fold. First, we investigate the robustness of routinel…
Constraining Chaos: Enforcing dynamical invariants in the training of recurrent neural networks
University of California San Diego（加利福尼亚大学圣迭戈分校）；Cooperative Institute for Research in Environmental Sciences（环境科学研究合作研究所）；University of Colorado Boulder（科罗拉多大学博尔德分校）；National Oceanic and Atmospheric Administration（美国国家海洋和大气管理局）；Scripps Institution of Oceanography（斯克里普斯海洋研究所）
Ecole Normale Supérieure（巴黎高等师范学院）；Université PSL（巴黎文理研究大学）；CNRS（法国国家科学研究中心）；Sorbonne Université（索邦大学）；Université de Paris（巴黎大学）；Collège de France（法兰西公学院）；LightOn（LightOn公司）；Ecole Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）
Reservoir Computing is a class of simple yet efficient Recurrent Neural Networks where internal weights are fixed at random and only a linear output layer is trained. In the large size limit, such random neural networks have a deep connection with kernel methods. Our contributions are threefold: a) We rigorously establish the recurrent kernel limit of Reservoir Computing and prove its convergence. b) We test our models on chaotic time series prediction, a classic but challenging benchmark in Reservoir Computing, and show how the Recurrent Kernel is competitive and computationally efficient whe…
University of California, Santa Cruz（加州大学圣克鲁兹分校）；University of California, Irvine（加州大学欧文分校）；Columbia University（哥伦比亚大学）
Recurrent Neural Networks (RNNs) frequently exhibit complicated dynamics, and their sensitivity to the initialization process often renders them notoriously hard to train. Recent works have shed light on such phenomena analyzing when exploding or vanishing gradients may occur, either of which is detrimental for training dynamics. In this paper, we point to a formal connection between RNNs and chaotic dynamical systems and prove a qualitatively stronger phenomenon about RNNs than what exploding gradients seem to suggest. Our main result proves that under standard initialization (e.g., He, Xavie…
ADAPT Research Centre（ADAPT研究中心）；Technological University Dublin（都柏林理工大学）
In order to build efficient deep recurrent neural architectures, it is essential to analyze the complexityof long distance dependencies (LDDs) of the dataset being modeled. In this paper, we presentdetailed analysis of the dependency decay curve exhibited by various datasets. The datasets sampledfrom a similar process (e.g. natural language, sequential MNIST, Strictlyk-Piecewise languages,etc) display variations in the properties of the dependency decay curve. Our analysis reveal thefactors resulting in these variations; such as (i) number of unique symbols in a dataset, (ii) size ofthe datase…
Mila, Université de Montréal（米拉-蒙特利尔大学）；Google Brain（谷歌大脑）；Microsoft Research（微软研究院）
Modelling long-term dependencies is a challenge for recurrent neural networks. This is primarily due to the fact that gradients vanish during training, as the sequence length increases. Gradients can be attenuated by transition operators and are attenuated or dropped by activation functions. Canonical architectures like LSTM alleviate this issue by skipping information through a memory mechanism. We propose a new recurrent architecture (Non-saturating Recurrent Unit; NRU) that relies on a memory mechanism but forgoes both saturating activation functions and saturating gates, in order to furthe…
The Pennsylvania State University（宾夕法尼亚州立大学）；Children’s Hospital of Philadelphia（费城儿童医院）
This work proposes a novel neural network architecture, called the Dynamically Controlled Recurrent Neural Network (DCRNN), specifically designed to model dynamical systems that are governed by ordinary differential equations (ODEs). The current state vectors of these types of dynamical systems only depend on their state-space models, along with the respective inputs and initial conditions. Long Short-Term Memory (LSTM) networks, which have proven to be very effective for memory-based tasks, may fail to model physical processes as they tend to memorize, rather than learn how to capture the inf…
École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；Yerevan State University（埃里温国立大学）
Reservoir Computing is a class of Recurrent Neural Networks with internal weights fixed at random. Stability relates to the sensitivity of the network state to perturbations. It is an important property in Reservoir Computing as it directly impacts performance. In practice, it is desirable to stay in a stable regime, where the effect of perturbations does not explode exponentially, but also close to the chaotic frontier where reservoir dynamics are rich. Open questions remain today regarding input regularization and discontinuous activation functions. In this work, we use the recurrent kernel …
Recurrent Neural Networks (RNNs) offer fast inference on long sequences but are hard to optimize and slow to train. Deep state-space models (SSMs) have recently been shown to perform remarkably well on long sequence modeling tasks, and have the added benefits of fast parallelizable training and RNN-like fast inference. However, while SSMs are superficially similar to RNNs, there are important differences that make it unclear where their performance boost over RNNs comes from. In this paper, we show that careful design of deep RNNs using standard signal propagation arguments can recover the imp…
Loyola Marymount University（洛约拉马利蒙特大学）；California State University, Long Beach（加州州立大学长滩分校）
We introduce an exceptionally simple gated recurrent neural network (RNN) that achieves performance comparable to well-known gated architectures, such as LSTMs and GRUs, on the word-level language modeling task. We prove that our model has simple, predicable and non-chaotic dynamics. This stands in stark contrast to more standard gated architectures, whose underlying dynamical systems exhibit chaotic behavior.
Indraprastha Institute of Information Technology, Delhi（德里英迪拉普拉斯信息技术学院）
In this paper, we train a recurrent neural network to learn dynamics of a chaotic road environment and to project the future of the environment on an image. Future projection can be used to anticipate an unseen environment for example, in autonomous driving. Road environment is highly dynamic and complex due to the interaction among traffic participants such as vehicles and pedestrians. Even in this complex environment, a human driver is efficacious to safely drive on chaotic roads irrespective of the number of traffic participants. The proliferation of deep learning research has shown the eff…
Surrogate testing techniques have been used widely to investigate the presence of dynamical nonlinearities, an essential ingredient of deterministic chaotic processes. Traditional surrogate testing subscribes to statistical hypothesis testing and investigates potential differences in discriminant statistics between the given empirical sample and its surrogate counterparts. The choice and estimation of the discriminant statistics can be challenging across short time series. Also, conclusion based on a single empirical sample is an inherent limitation. The present study proposes a recurrent neur…
Princeton University（普林斯顿大学）；Institute for Advanced Study（高等研究院）；CUNY（纽约市立大学）
Recurrent neural networks (RNNs) are powerful dynamical models, widely used in machine learning (ML) and neuroscience. Prior theoretical work has focused on RNNs with additive interactions. However, gating - i.e. multiplicative - interactions are ubiquitous in real neurons and also the central feature of the best-performing RNNs in ML. Here, we show that gating offers flexible control of two salient features of the collective dynamics: i) timescales and ii) dimensionality. The gate controlling timescales leads to a novel, marginally stable state, where the network functions as a flexible integ…
ETIS UMR 8051（ETIS UMR 8051实验室）；CY University（塞尔吉-巴黎大学）；ENSEA（法国国立高等电子及应用工程师学院）；CNRS（法国国家科学研究中心）
As a phenomenon in dynamical systems allowing autonomous switching between stable behaviors, chaotic itinerancy has gained interest in neurorobotics research. In this study, we draw a connection between this phenomenon and the predictive coding theory by showing how a recurrent neural network implementing predictive coding can generate neural trajectories similar to chaotic itinerancy in the presence of input noise. We propose two scenarios generating random and past-independent attractor switching trajectories using our model.
University of Potsdam（波茨坦大学）；Vrije Universiteit Amsterdam（阿姆斯特丹自由大学）；Ambrosys GmbH（Ambrosys有限公司）
We investigate the predictive power of recurrent neural networks for oscillatory systems not only on the attractor, but in its vicinity as well. For this we consider systems perturbed by an external force. This allows us to not merely predict the time evolution of the system, but also study its dynamical properties, such as bifurcations, dynamical response curves, characteristic exponents etc. It is shown that they can be effectively estimated even in some regions of the state space where no input data were given. We consider several different oscillatory examples, including self-sustained, ex…
Emergent behaviors are in the focus of recent research interest. It is then of considerable importance to investigate what optimizations suit the learning and prediction of chaotic systems, the putative candidates for emergence. We have compared L1 and L2 regularizations on predicting chaotic time series using linear recurrent neural networks. The internal representation and the weights of the networks were optimized in a unifying framework. Computational tests on different problems indicate considerable advantages for the L1 regularization: It had considerably better learning time and better …
It is a known fact that training recurrent neural networks for tasks that have long term dependencies is challenging. One of the main reasons is the vanishing or exploding gradient problem, which prevents gradient information from propagating to early layers. In this paper we propose a simple recurrent architecture, the Fourier Recurrent Unit (FRU), that stabilizes the gradients that arise in its training while giving us stronger expressive power. Specifically, FRU summarizes the hidden states $h^{(t)}$ along the temporal dimension with Fourier basis functions. This allows gradients to easily …
University of Tehran（德黑兰大学）；KTH Royal Institute of Technology（瑞典皇家理工学院）；Swedish e-Science Research Centre (SeRC)（瑞典e-Science研究中心）
The capabilities of recurrent neural networks and Koopman-based frameworks are assessed in the prediction of temporal dynamics of the low-order model of near-wall turbulence by Moehlis et al. (New J. Phys. 6, 56, 2004). Our results show that it is possible to obtain excellent reproductions of the long-term statistics and the dynamic behavior of the chaotic system with properly trained long-short-term memory (LSTM) networks, leading to relative errors in the mean and the fluctuations below $1\%$. Besides, a newly developed Koopman-based framework, called Koopman with nonlinear forcing (KNF), le…
Traditional Recurrent Neural Networks assume vectorized data as inputs. However many data from modern science and technology come in certain structures such as tensorial time series data. To apply the recurrent neural networks for this type of data, a vectorisation process is necessary, while such a vectorisation leads to the loss of the precise information of the spatial or longitudinal dimensions. In addition, such a vectorized data is not an optimum solution for learning the representation of the longitudinal data. In this paper, we propose a new variant of tensorial neural networks which d…
Faculty of Computer Science and Engineering The Hebrew University of Jerusalem（计算机科学与工程学院 希伯来大学耶路撒冷分校）；The Hebrew University of Jerusalem（希伯来大学耶路撒冷分校）；Faculty of Computer Science and Engineering（计算机科学与工程学院）
A key attribute that drives the unprecedented success of modern Recurrent Neural Networks (RNNs) on learning tasks which involve sequential data, is their ability to model intricate long-term temporal dependencies. However, a well established measure of RNNs long-term memory capacity is lacking, and thus formal understanding of the effect of depth on their ability to correlate data throughout time is limited. Specifically, existing depth efficiency results on convolutional networks do not suffice in order to account for the success of deep RNNs on data of varying lengths. In order to address t…
The design of recurrent neural networks (RNNs) to accurately process sequential inputs with long-time dependencies is very challenging on account of the exploding and vanishing gradient problem. To overcome this, we propose a novel RNN architecture which is based on a structure preserving discretization of a Hamiltonian system of second-order ordinary differential equations that models networks of oscillators. The resulting RNN is fast, invertible (in time), memory efficient and we derive rigorous bounds on the hidden state gradients to prove the mitigation of the exploding and vanishing gradi…
Central Institute of Mental Health（中央心理健康研究所）；Heidelberg University（海德堡大学）
Recurrent neural networks (RNNs) are wide-spread machine learning tools for modeling sequential and time series data. They are notoriously hard to train because their loss gradients backpropagated in time tend to saturate or diverge during training. This is known as the exploding and vanishing gradient problem. Previous solutions to this issue either built on rather complicated, purpose-engineered architectures with gated memory buffers, or - more recently - imposed constraints that ensure convergence to a fixed point or restrict (the eigenspectrum of) the recurrence matrix. Such constraints, …
Mechanical and Aerospace Engineering, Princeton University（普林斯顿大学机械与航空航天工程系（注：此处按要求提取到大学级别，故保留系名与大学名））；Princeton University（普林斯顿大学）
This paper describes a method for learning low-dimensional approximations of nonlinear dynamical systems, based on neural-network approximations of the underlying Koopman operator. Extended Dynamic Mode Decomposition (EDMD) provides a useful data-driven approximation of the Koopman operator for analyzing dynamical systems. This paper addresses a fundamental problem associated with EDMD: a trade-off between representational capacity of the dictionary and over-fitting due to insufficient data. A new neural network architecture combining an autoencoder with linear recurrent dynamics in the encode…
LSTM or Long Short Term Memory Networks is a specific type of Recurrent Neural Network (RNN) that is very effective in dealing with long sequence data and learning long term dependencies. In this work, we perform sentiment analysis on a GOP Debate Twitter dataset. To speed up training and reduce the computational cost and time, six different parameter reduced slim versions of the LSTM model (slim LSTM) are proposed. We evaluate two of these models on the dataset. The performance of these two LSTM models along with the standard LSTM model is compared. The effect of Bidirectional LSTM Layers is …
Learning long-term dependencies is a key long-standing challenge of recurrent neural networks (RNNs). Hierarchical recurrent neural networks (HRNNs) have been considered a promising approach as long-term dependencies are resolved through shortcuts up and down the hierarchy. Yet, the memory requirements of Truncated Backpropagation Through Time (TBPTT) still prevent training them on very long sequences. In this paper, we empirically show that in (deep) HRNNs, propagating gradients back from higher to lower levels can be replaced by locally computable losses, without harming the learning capabil…
We introduce a data-driven forecasting method for high-dimensional chaotic systems using long short-term memory (LSTM) recurrent neural networks. The proposed LSTM neural networks perform inference of high-dimensional dynamical systems in their reduced order space and are shown to be an effective set of nonlinear approximators of their attractor. We demonstrate the forecasting performance of the LSTM and compare it with Gaussian processes (GPs) in time series obtained from the Lorenz 96 system, the Kuramoto-Sivashinsky equation and a prototype climate model. The LSTM networks outperform the GP…
Recommender systems objectives can be broadly characterized as modeling user preferences over short-or long-term time horizon. A large body of previous research studied long-term recommendation through dimensionality reduction techniques applied to the historical user-item interactions. A recently introduced session-based recommendation setting highlighted the importance of modeling short-term user preferences. In this task, Recurrent Neural Networks (RNN) have shown to be successful at capturing the nuances of user's interactions within a short time window. In this paper, we evaluate RNN-base…
Laboratoire de Recherche en Informatique（计算机研究实验室）；Facebook Artificial Intelligence Research（Facebook人工智能研究院）；Université Paris Sud（巴黎南大学）
Successful recurrent models such as long short-term memories (LSTMs) and gated recurrent units (GRUs) use ad hoc gating mechanisms. Empirically these models have been found to improve the learning of medium to long term temporal dependencies and to help with vanishing gradient issues. We prove that learnable gates in a recurrent model formally provide quasi- invariance to general time transformations in the input data. We recover part of the LSTM architecture from a simple axiomatic approach. This result leads to a new way of initializing gate biases in LSTMs and GRUs. Ex- perimentally, this n…
Recurrent neural network is a powerful model that learns temporal patterns in sequential data. For a long time, it was believed that recurrent networks are difficult to train using simple optimizers, such as stochastic gradient descent, due to the so-called vanishing gradient problem. In this paper, we show that learning longer term patterns in real data, such as in natural language, is perfectly possible using gradient descent. This is achieved by using a slight structural modification of the simple recurrent neural network architecture. We encourage some of the hidden units to change their s…
In this study we demonstrate a novel Brain Computer Interface (BCI) approach to detect driver distraction events to improve road safety. We use a commercial wireless headset that generates EEG signals from the brain. We collected real EEG signals from participants who undertook a 40-minute driving simulation and were required to perform different tasks while driving. These signals are segmented into short windows and labelled using a time series classification (TSC) model. We studied different TSC approaches and designed a Long-term Recurrent Convolutional Network (LCRN) model for this task. O…
Recurrent neural networks (RNNs) provide state-of-the-art performances in a wide variety of tasks that require memory. These performances can often be achieved thanks to gated recurrent cells such as gated recurrent units (GRU) and long short-term memory (LSTM). Standard gated cells share a layer internal state to store information at the network level, and long term memory is shaped by network-wide recurrent connection weights. Biological neurons on the other hand are capable of holding information at the cellular level for an arbitrary long amount of time through a process called bistability…
Safran Electronics & Defense（赛峰电子与防务公司）；AIM, CEA, CNRS, Université Paris-Saclay, Université de Paris（AIM实验室，法国原子能与替代能源委员会，法国国家科学研究中心，巴黎萨克雷大学，巴黎大学）；DR2I, Institut Polytechnique des Sciences Avancées（DR2I实验室，先进科学理工学院）
Recurrent models have gained popularity in deep learning (DL) based video super-resolution (VSR), due to their increased computational efficiency, temporal receptive field and temporal consistency compared to sliding-window based models. However, when inferring on long video sequences presenting low motion (i.e. in which some parts of the scene barely move), recurrent models diverge through recurrent processing, generating high frequency artifacts. To the best of our knowledge, no study about VSR pointed out this instability problem, which can be critical for some real-world applications. Vide…
Reinforcement learning (RL) agents performing complex tasks must be able to remember observations and actions across sizable time intervals. This is especially true during the initial learning stages, when exploratory behaviour can increase the delay between specific actions and their effects. Many new or popular approaches for learning these distant correlations employ backpropagation through time (BPTT), but this technique requires storing observation traces long enough to span the interval between cause and effect. Besides memory demands, learning dynamics like vanishing gradients and slow …
Recurrent neural networks (RNN) are at the core of modern automatic speech recognition (ASR) systems. In particular, long-short term memory (LSTM) recurrent neural networks have achieved state-of-the-art results in many speech recognition tasks, due to their efficient representation of long and short term dependencies in sequences of inter-dependent features. Nonetheless, internal dependencies within the element composing multidimensional features are weakly considered by traditional real-valued representations. We propose a novel quaternion long-short term memory (QLSTM) recurrent neural netw…
Munich Data Science Institute（慕尼黑数据科学研究所）；School of Computation, Information and Technology, Technical University of Munich（技术大学慕尼黑计算、信息与技术学院）；Interdisciplinary Center for Scientific Computing (IWR)（跨学科科学计算中心）；Department of Mathematics and Computer Science, Heidelberg University（海德堡大学数学与计算机科学系）
The vanishing and exploding gradient problems are well-studied obstacles that make it difficult for recurrent neural networks to learn long-term time dependencies. We propose a reparameterization of standard recurrent neural networks to update linear transformations in a provably norm-preserving way through Givens rotations. Additionally, we use the absolute value function as an element-wise non-linearity to preserve the norm of backpropagated signals over the entire network. We show that this reparameterization reduces the number of parameters and maintains the same algorithmic complexity as …
Optimal training of finitely-sampled quantum reservoir computers for forecasting of chaotic dynamics
Imperial College London（伦敦帝国理工学院）；City St George’s, University of London（伦敦大学城市圣乔治学院）；The Alan Turing Institute（艾伦·图灵研究所）；Politecnico di Torino（都灵理工大学）
Reduced Order Modeling with Shallow Recurrent Decoder Networks
Department of Mechanical Engineering, Politecnico di Milano, Milano, Italy（米兰理工学院机械工程系）；Department of Mechanical Engineering, University of Washington, Seattle, WA（华盛顿大学机械工程系）；MOX - Department of Mathematics, Politecnico di Milano, Milano, Italy（米兰理工学院数学系）；Department of Applied Mathematics and Electrical and Computer Engineering, University of Washington, Seattle, WA（华盛顿大学应用数学与电气与计算机工程系）
A key attribute that drives the unprecedented success of modern Recurrent Neural Networks (RNNs) on learning tasks which involve sequential data, is their ability to model intricate long-term temporal dependencies. However, a well established measure of RNNs long-term memory capacity is lacking, and thus formal understanding of the effect of depth on their ability to correlate data throughout time is limited. Specifically, existing depth efficiency results on convolutional networks do not suffice in order to account for the success of deep RNNs on data of varying lengths. In order to address t…
We forecast two different chaotic dynamics of the quasiperiodically forced logistic map using the well-known deep learning framework Long Short-Term Memory. We generate two data sets and use one in the training process and the other in the testing process. The predicted values are evaluated using the metric called Root Mean Square Error and visualized using the scatter plots. The robustness of the Long Short-Term Memory model is evaluated using the number of units in the layers of the model. We also make multi-step forecasting of the considered system. We show that the considered Long Short-Te…
Schmalkalden University of Applied Sciences（施马尔卡尔登应用科学大学）；Singapore University of Technology and Design（新加坡科技设计大学）
Long Short-Term Memory Recurrent Neural Networks (LSTM-RNN) are one of the most powerful dynamic classifiers publicly known. The network itself and the related learning algorithms are reasonably well documented to get an idea how it works. This paper will shed more light into understanding how LSTM-RNNs evolved and why they work impressively well, focusing on the early, ground-breaking publications. We significantly improved documentation and fixed a number of errors and inconsistencies that accumulated in previous publications. To support understanding we as well revised and unified the notat…
University of Electronic Science and Technology of China（电子科技大学）；The University of Queensland（昆士兰大学）
Recurrent neural networks (RNNs) are widely used as a memory model for sequence-related problems. Many variants of RNN have been proposed to solve the gradient problems of training RNNs and process long sequences. Although some classical models have been proposed, capturing long-term dependence while responding to short-term changes remains a challenge. To this problem, we propose a new model named Dual Recurrent Neural Networks (DuRNN). The DuRNN consists of two parts to learn the short-term dependence and progressively learn the long-term dependence. The first part is a recurrent neural netw…
