Gianniotis, Nikolaos, Tino, Peter, Spreckley, Steve, Raychaudhury, Somak
We present a probabilistic generative approach for constructing topographic maps of light curves from eclipsing binary stars. The model defines a low-dimensional manifold of local noise models...
Cuevas-Tello, Juan C., Tino, Peter, Raychaudhury, Somak, Yao, Xin, Harva, Markus
We study the problem of estimating the time delay between two signals representing delayed, irregularly sampled and noisy versions of the same underlying pattern. We propose and demonstrate an...
Estimating Time Delay in Gravitationally Lensed Fluxes (2009)
Tino, Peter, Cuevas-Tello, Juan C., Raychaudhury, Somak
We study the problem of estimating the time delay between two signals representing delayed, irregularly sampled and noisy versions of the same underlying pattern. We propose a kernel-based technique...
Jin Li, Xiaoli Li, Colin Frayn, Peter Tino, Xin Yao
Applications (CERCIA) is a unique new initiative, aimed to be an international leader in applied research and knowledge transfer of computational intelligence techniques for the benefit of industry...
Topographic Organization of User Preference Patterns in Collaborative Filtering (2008)
Peter Tino, Gabriela Polcicova
We introduce topographic versions of two latent class models for collaborative filtering.
Optimization dynamics using self-organizing neural networks (SONN) driven by softmax weight renormalization has been shown to be capable of intermittent search for high-quality solutions in...
Peter Tino, Michal Stancík, Lubica Benusková
. We use a recurrent version of the Bienenstock, Cooper and Munro network (RBCMN) with lateral inhibition [2] to map histories of symbols into activations in the recurrent layer. After training the...
Semi-Supervised Construction of General Visualization Hierarchies (2007)
Peter Tino, Yi Sun, Ian Nabney
Abstract We have recently developed a principled approach to interactive non-linear hierarchical visualization [8] based on the Generative Topographic Mapping (GTM). Hierarchical plots are needed...
Tsungnan Lin, Bill G. Horne, Peter Tino
long--term dependencies is not as difficult with NARX recurrent neural networks
Peter Tino, Bill G. Horne, C. Lee Giles
The position, number and stability types of fixed points of a two--neuron recurrent network with nonzero weights are investigated. Using simple geometrical arguments in the space of derivatives of...
Christian Schittenkopf, Peter Tino, Georg Dorffner
Essentially, there are two notions of volatility in literature: historical volatility and implied volatility. While measures of the former notion are derived from historical returns by (weighted)...
Building Predictive Models on Complex (2007)
Symbolic Sequences Via, Peter Tino
We use a recurrent version of the Bienenstock, Cooper and Munro network (RBCMN) with lateral inhibition [2] to map histories of symbols into activations in the recurrent layer. After training the...
How accurate are the time delay estimates in gravitational lensing? (2006)
Cuevas-Tello, Juan C., Tino, Peter, Raychaudhury, Somak
We present a novel approach to estimate the time delay between light curves of multiple images in a gravitationally lensed system, based on Kernel methods in the context of machine learning. We...
Managing Diversity in Regression Ensembles (2005)
Brown, Gavin, Wyatt, Jeremy, Tino, Peter
Ensembles are a widely used and effective technique in machine learning---their success is commonly attributed to the degree of disagreement, or 'diversity', within the ensemble. For ensembles where...
Semisupervised learning of hierarchical latent trait models for data visualisation (2005)
Ian T. Nabney, Yi Sun, Peter Tino, Ata Kabán
Abstract—Recently, we have developed the hierarchical Generative Topographic Mapping (HGTM), an interactive method for visualization of large high-dimensional real-valued data sets. In this paper,...
Markovian architectural bias of recurrent neural networks (2002)
We have recently shown that when initialized with "small " weights, recurrent neural networks (RNNs) with standard sigmoid-type activation functions are inherently biased towards...
Peter Tino, Bill G. Horne, C. Lee Giles
The position, number and stability types of fixed points of a two--neuron recurrent network with nonzero weights are investigated. Using geometrical arguments in the space of derivatives of the...
The benefit of information reduction for trading strategies (2000)
Schittenkopf, Christian, Tino, Peter, Dorffner, Georg
Motivated by previous findings that discretization of financial time series can effectively filter the data and reduce the noise, this experimental study compares the trading performance of...
Tino, Peter, Schittenkopf, Christian, Dorffner, Georg
In this paper we investigate the potential of the analysis of noisy non-stationary time series by quantizing it into streams of discrete symbols and applying finite-memory symbolic predictors. The...
The profitability of trading volatility using real-valued and symbolic models (2000)
Christian Schittenkopf, Peter Tino, Georg Dorffner
Essentially, there are two notions of volatility in literature: historical volatility and implied volatility. While measures of the former notion are derived from historical returns by (weighted)...
A Symbolic dynamics approach to volatility prediction (2000)
Peter Tino, Christian Schittenkopf, Georg Dorffner, Engelbert J. Dockner
We consider the problem of predicting the direction of daily volatility changes in the Dow Jones Industrial Average (DJIA). This is accomplished by quantizing a series of historic volatility changes...
Peter Tino, Christian Schittenkopf, Georg Dorffner
In this paper we investigate the potential of the analysis of noisy non-stationary time series by quantizing it into streams of discrete symbols and applying finitememory symbolic predictors. The...
The Benefit of Information Reduction for Trading Strategies (2000)
Christian Schittenkopf, Peter Tino, Georg Dorffner
Motivated by previous findings that discretization of financial time series can effectively filter the data and reduce the noise, this experimental study compares the trading performance of...
Peter Tino, Michal Stancík, Lubica Benusková
. Activation patterns across recurrent units in recurrent neural networks (RNNs) can be thought of as spatial codes of the history of inputs seen so far. When trained on symbolic sequences to perform...
Peter Tino, Bill Horne, C. Lee Giles
We perform a detailed xed-point analysis of two-unit recurrent neural networks with sigmoid-shaped transfer functions. Using geometrical arguments in the space of transfer function derivatives, we...
While much work has been done in neural based modeling of real valued chaotic time series, little effort has been devoted to address similar problems in the symbolic domain. We investigate the...
Spatial representation of symbolic sequences through iterative function systems (1999)
Abstract--- Jeffrey proposed a graphic representation of DNA sequences using Barnsley's iterative function systems. In spite of further developments in this direction, the proposed graphic...
A Symbolic Dynamics Approach to Volatility Prediction (1999)
Peter Tino, Christian Schittenkopf, Georg Dorffner, Engelbert J. Dockner
We consider the problem of predicting the direction of daily volatility changes in the Dow Jones Industrial Average (DJIA). This is accomplished by quantizing a series of historic volatility changes...
Recurrent neural networks with iterated function systems dynamics (1998)
We suggest a recurrent neural network (RNN) model with a recurrent part corresponding to iterative function systems (IFS) introduced by Barnsley [1] as a fractal image compression mechanism. The key...
Constructing finite-context sources from fractal representations of symbolic sequences (1998)
We propose a novel approach to constructing predictive models on long complex symbolic sequences. The models are constructed by first transforming the training sequence n-block structure into a...
Spatial representation of symbolic sequences through iterative function systems (1998)
Jeffrey proposed a graphic representation of DNA sequences using Barnsley's iterative function systems. In spite of further developments in this direction (Oliver et. al, 1993), (Roman-Roldan et. al,...
A symbolic dynamics approach to volatility prediction (1998)
Tino, Peter, Schittenkopf, Christian, Dorffner, Georg, Dockner, Engelbert J.
We consider the problem of predicting the direction of daily volatility changes in the Dow Jones Industrial Average (DJIA). This is accomplished by quantizing a series of historic volatility changes...
Spatial representation of symbolic sequences through iterative function systems (1998)
Jeffrey proposed a graphic representation of DNA sequences using Barnsley's iterative function systems. In spite of further developments in this direction (Oliver et. al, 1993), (Roman-Roldan et. al,...
Constructing finite-context sources from fractal representations of symbolic sequences (1998)
We propose a novel approach to constructing predictive models on long complex symbolic sequences. The models are constructed by first transforming the training sequence n-block structure into a...
Recurrent neural networks with iterated function systems dynamics (1998)
We suggest a recurrent neural network (RNN) model with a recurrent part corresponding to iterative function systems (IFS) introduced by Barnsley [1] as a fractal image compression mechanism. The key...
We train recurrent neural network on a single, long, complex symbolic sequence with positive entropy. Training process is monitored through information theory based performance measures. We show that...
Recurrent neural networks with Iterated Function Systems dynamics (1998)
We suggest a recurrent neural network (RNN) model with a recurrent part corresponding to iterative function systems (IFS) introduced by Barnsley [1] as a fractal image compression mechanism. The key...
Constructing Finite-Context Sources From Fractal Representations of Symbolic Sequences (1998)
We propose a novel approach to constructing predictive models on long complex symbolic sequences. The models are constructed by first transforming the training sequence n-block structure into a...
Spatial Representation of Symbolic Sequences through Iterative Function Systems (1998)
Jeffrey [10] proposed a graphic representation of DNA sequences using Barnsley 's iterative function systems. In spite of further developments in this direction [19, 25, 13], the proposed...
Spatial Representation of Symbolic Sequences through Iterative Function Systems (1998)
Peter Tino Austrian, Peter Tino
Jeffrey [10] proposed a graphic representation of DNA sequences using Barnsley 's iterative function systems. In spite of further developments in this direction [19, 25, 13], the proposed...
Bill G. Horne, C. Lee Giles, Pete C. Collingwood, School Of Computing, Man Sci, Peter Tino, ...
We present two approaches to the analysis of the relationship between a recurrent neural network (RNN) and the finite state machine M the network is able to exactly mimic. First, the network is...
Constructing Finite-Context Sources From Fractal Representations of Symbolic Sequences (1998)
We propose a novel approach to constructing predictive models on long complex symbolic sequences. The models are constructed by first transforming the training sequence n-block structure into a...
Self Organizing Map for Fast Cluster Detection in High-Dimensional Spaces (1997)
We study the problem of fast detection of dense, sufficiently separated clusters in (possibly) high dimensional spaces using self-organizing maps. To this end, we formulate the self-organizing map as...
Learning long--term dependencies in NARX recurrent neural networks (1996)
Tsungnan Lin Bill, Bill G. Horne, Peter Tino
It has recently been shown that gradient-descent learning algorithms for recurrent neural networks can perform poorly on tasks that involve long--term dependencies, i.e. those problems for which the...
Learning long-term dependencies in NARX recurrent neural networks (1996)
Tsungnan Lin, Bill G. Horne, Peter Tino, C. Lee Giles
It has recently been shown that gradient-descent learning algorithms for recurrent neural networks can perform poorly on tasks that involve long--term dependencies, i.e. those problems for which the...
Learning long--term dependencies is not as difficult with NARX recurrent neural networks (1996)
Tsungnan Lin, Bill G. Horne, Peter Tino
It has recently been shown that gradient descent learning algorithms for recurrent neural networks can perform poorly on tasks that involve long--term dependencies, i.e. those problems for which the...
Modeling Complex Symbolic Sequences with Neural and Hybrid Neural Based Systems (1996)
While much work has been done in neural based modeling of real valued chaotic time series, little effort has been devoted to address similar problems in the symbolic domain. In this paper we (1)...
Learning Long-Term Dependencies is Not as Difficult With NARX Recurrent Neural Networks (1995)
Lin, Tsungnan, Horne, Bill G., Tino, Peter, Giles, C. Lee
It has recently been shown that gradient descent learning algorithms for recurrent neural networks can perform poorly on tasks that involve long- term dependencies, i.e. those problems for which the...
Learning Long-Term Dependencies is Not as Difficult With NARX Recurrent Neural Networks (1995)
Lin, Tsungnan, Horne, Bill G., Tino, Peter, Giles, C. Lee
It has recently been shown that gradient descent learning algorithms for recurrent neural networks can perform poorly on tasks that involve long- term dependencies, i.e. those problems for which the...
Tino, Peter, Horne, Bill G., Giles, C. Lee
The position, number and stability types of fixed points of a two--neuron recurrent network with nonzero weights are investigated. Using simple geometrical arguments in the space of derivatives of...
Tino, Peter, Horne, Bill G., Giles, C. Lee
The position, number and stability types of fixed points of a two--neuron recurrent network with nonzero weights are investigated. Using simple geometrical arguments in the space of derivatives of...
Tino, Peter, Horne, Bill G., Giles, C. Lee
We present two approaches to the analysis of the relationship between a recurrent neural network (RNN) and the finite state machine \( {\cal M} \) the network is able to exactly mimic. First, the...
Tino, Peter, Horne, Bill G., Giles, C. Lee
We present two approaches to the analysis of the relationship between a recurrent neural network (RNN) and the finite state machine \( {\cal M} \) the network is able to exactly mimic. First, the...
Learning and extracting initial mealy automata with a modular neural network model (1995)
A hybrid recurrent neural network is shown to learn small initial mealy machines (that can be thought of as translation machines translating input strings to corresponding output strings, as opposed...
Peter Tino, Bill G. Horne, C. Lee Giles
The position, number and stability types of fixed points of a two--neuron recurrent network with nonzero weights are investigated. Using simple geometrical arguments in the space of derivatives of...
The Benefit of Information Reduction for Trading Strategies.
Schittenkopf, Christian, Tino, Peter, Dorffner, Georg
Motivated by previous findings that discretization of financial time series can effectively filter the data and reduce the noise, this experimental study compares the trading performance of...
Does money matter in inflation forecasting?
Jane M. Binner, Peter Tino, Jonathan Tepper, Richard G. Anderson, Barry Jones, Graham Kendall
This paper provides the most fully comprehensive evidence to date on whether or not monetary aggregates are valuable for forecasting US inflation in the early to mid 2000s. We explore a wide range of...
Does money matter in inflation forecasting?
Jane M. Binner, Peter Tino, Jonathan Tepper, Richard G. Anderson, Barry Jones, Graham Kendall
This paper provides the most fully comprehensive evidence to date on whether or not monetary aggregates are valuable for forecasting US inflation in the early to mid 2000s. We explore a wide range of...