Hierarchical Prediction Machines and Big Data Analytics

An emerging consensus in cognitive science views the biological brain as a hierarchically-organized predictive processing system. This is a system in which higher-order regions are continuously attempting to predict the activity of lower-order regions at a variety of (increasingly abstract) spatial and temporal scales. The brain is thus revealed as a hierarchical prediction machine that is constantly engaged in the effort to predict the flow of information originating from the sensory surfaces. Such a view seems to afford a great deal of explanatory leverage when it comes to a broad swathe of seemingly disparate psychological phenomena (e.g., learning, memory, perception, action, emotion, planning, reason, imagination, and conscious experience). In the most positive case, the predictive processing story seems to provide our first glimpse at what a unified (computationally-tractable and neurobiological plausible) account of human psychology might look like. This obviously marks out one reason why such models should be the focus of current empirical and theoretical attention. Another reason, however, is rooted in the potential of such models to advance the current state-of-the-art in machine intelligence and machine learning. Interestingly, the vision of the brain as a hierarchical prediction machine is one that establishes contact with work that goes under the heading of 'deep learning'. Deep learning systems thus often attempt to make use of predictive processing schemes and (increasingly abstract) generative models as a means of supporting the analysis of large data sets. But are such computational systems sufficient (by themselves) to provide a route to general human-level analytic capabilities? I will argue that they are not and that closer attention to a broader range of forces and factors (many of which are not confined to the neural realm) may be required to understand what it is that gives human cognition its distinctive (and largely unique) flavour. The vision that emerges is one of 'homomimetic deep learning systems', systems that situate a hierarchically-organized predictive processing core within a larger nexus of developmental, behavioural, symbolic, technological and social influences. Relative to that vision, I suggest that we should see the Web as a form of 'cognitive ecology', one that is as much involved with the transformation of machine intelligence as it is with the progressive reshaping of our own cognitive capabilities.

Data assurance in opaque computations

The chess endgame is increasingly being seen through the lens of, and therefore effectively defined by, a data ‘model’ of itself. It is vital that such models are clearly faithful to the reality they purport to represent. This paper examines that issue and systems engineering responses to it, using the chess endgame as the exemplar scenario. A structured survey has been carried out of the intrinsic challenges and complexity of creating endgame data by reviewing the past pattern of errors during work in progress, surfacing in publications and occurring after the data was generated. Specific measures are proposed to counter observed classes of error-risk, including a preliminary survey of techniques for using state-of-the-art verification tools to generate EGTs that are correct by construction. The approach may be applied generically beyond the game domain.

Chess endgames: 6-man data and strategy

While Nalimov’s endgame tables for Western Chess are the most used today, their Depth-to-Mate metric is not the most efficient or effective in use. The authors have developed and used new programs to create tables to alternative metrics and recommend better strategies for endgame play.

A new linear initialization in SOM for biomolecular data

In the past decade, the amount of data in biological field has become larger and larger; Bio-techniques for analysis of biological data have been developed and new tools have been introduced. Several computational methods are based on unsupervised neural network algorithms that are widely used for multiple purposes including clustering and visualization, i.e. the Self Organizing Maps (SOM). Unfortunately, even though this method is unsupervised, the performances in terms of quality of result and learning speed are strongly dependent from the neuron weights initialization. In this paper we present a new initialization technique based on a totally connected undirected graph, that report relations among some intersting features of data input. Result of experimental tests, where the proposed algorithm is compared to the original initialization techniques, shows that our technique assures faster learning and better performance in terms of quantization error.

Automated pre-processing strategies for species occurrence data used in biodiversity modelling

To construct Biodiversity richness maps from Environmental Niche Models (ENMs) of thousands of species is time consuming. A separate species occurrence data pre-processing phase enables the experimenter to control test AUC score variance due to species dataset size. Besides, removing duplicate occurrences and points with missing environmental data, we discuss the need for coordinate precision, wide dispersion, temporal and synonymity filters. After species data filtering, the final task of a pre-processing phase should be the automatic generation of species occurrence datasets which can then be directly ’plugged-in’ to the ENM. A software application capable of carrying out all these tasks will be a valuable time-saver particularly for large scale biodiversity studies.

Surface based hypothesis verification in intensity images using geometric and appearance data

In this paper we discuss current work concerning Appearance-based and CAD-based vision; two opposing vision strategies. CAD-based vision is geometry based, reliant on having complete object centred models. Appearance-based vision builds view dependent models from training images. Existing CAD-based vision systems that work with intensity images have all used one and zero dimensional features, for example lines, arcs, points and corners. We describe a system we have developed for combining these two strategies. Geometric models are extracted from a commercial CAD library of industry standard parts. Surface appearance characteristics are then learnt automatically by observing actual object instances. This information is combined with geometric information and is used in hypothesis evaluation. This augmented description improves the systems robustness to texture, specularities and other artifacts which are hard to model with geometry alone, whilst maintaining the advantages of a geometric description.

Generation, description and storage of dendritic morphology data

It is generally assumed that the variability of neuronal morphology has an important effect on both the connectivity and the activity of the nervous system, but this effect has not been thoroughly investigated. Neuroanatomical archives represent a crucial tool to explore structure–function relationships in the brain. We are developing computational tools to describe, generate, store and render large sets of three–dimensional neuronal structures in a format that is compact, quantitative, accurate and readily accessible to the neuroscientist. Single–cell neuroanatomy can be characterized quantitatively at several levels. In computer–aided neuronal tracing files, a dendritic tree is described as a series of cylinders, each represented by diameter, spatial coordinates and the connectivity to other cylinders in the tree. This ‘Cartesian’ description constitutes a completely accurate mapping of dendritic morphology but it bears little intuitive information for the neuroscientist. In contrast, a classical neuroanatomical analysis characterizes neuronal dendrites on the basis of the statistical distributions of morphological parameters, e.g. maximum branching order or bifurcation asymmetry. This description is intuitively more accessible, but it only yields information on the collective anatomy of a group of dendrites, i.e. it is not complete enough to provide a precise ‘blueprint’ of the original data. We are adopting a third, intermediate level of description, which consists of the algorithmic generation of neuronal structures within a certain morphological class based on a set of ‘fundamental’, measured parameters. This description is as intuitive as a classical neuroanatomical analysis (parameters have an intuitive interpretation), and as complete as a Cartesian file (the algorithms generate and display complete neurons). The advantages of the algorithmic description of neuronal structure are immense. If an algorithm can measure the values of a handful of parameters from an experimental database and generate virtual neurons whose anatomy is statistically indistinguishable from that of their real counterparts, a great deal of data compression and amplification can be achieved. Data compression results from the quantitative and complete description of thousands of neurons with a handful of statistical distributions of parameters. Data amplification is possible because, from a set of experimental neurons, many more virtual analogues can be generated. This approach could allow one, in principle, to create and store a neuroanatomical database containing data for an entire human brain in a personal computer. We are using two programs, L–NEURON and ARBORVITAE, to investigate systematically the potential of several different algorithms for the generation of virtual neurons. Using these programs, we have generated anatomically plausible virtual neurons for several morphological classes, including guinea pig cerebellar Purkinje cells and cat spinal cord motor neurons. These virtual neurons are stored in an online electronic archive of dendritic morphology. This process highlights the potential and the limitations of the ‘computational neuroanatomy’ strategy for neuroscience databases.

The Climate-G testbed: towards large scale distributed data management for climate change

Climate-G is a large scale distributed testbed devoted to climate change research. It is an unfunded effort started in 2008 and involving a wide community both in Europe and US. The testbed is an interdisciplinary effort involving partners from several institutions and joining expertise in the field of climate change and computational science. Its main goal is to allow scientists carrying out geographical and cross-institutional data discovery, access, analysis, visualization and sharing of climate data. It represents an attempt to address, in a real environment, challenging data and metadata management issues. This paper presents a complete overview about the Climate-G testbed highlighting the most important results that have been achieved since the beginning of this project.

Distributed classification for pocket data mining

Distributed and collaborative data stream mining in a mobile computing environment is referred to as Pocket Data Mining PDM. Large amounts of available data streams to which smart phones can subscribe to or sense, coupled with the increasing computational power of handheld devices motivates the development of PDM as a decision making system. This emerging area of study has shown to be feasible in an earlier study using technological enablers of mobile software agents and stream mining techniques [1]. A typical PDM process would start by having mobile agents roam the network to discover relevant data streams and resources. Then other (mobile) agents encapsulating stream mining techniques visit the relevant nodes in the network in order to build evolving data mining models. Finally, a third type of mobile agents roam the network consulting the mining agents for a final collaborative decision, when required by one or more users. In this paper, we propose the use of distributed Hoeffding trees and Naive Bayes classifers in the PDM framework over vertically partitioned data streams. Mobile policing, health monitoring and stock market analysis are among the possible applications of PDM. An extensive experimental study is reported showing the effectiveness of the collaborative data mining with the two classifers.

Homogeneous and heterogeneous distributed classification for pocket data mining

Pocket Data Mining (PDM) describes the full process of analysing data streams in mobile ad hoc distributed environments. Advances in mobile devices like smart phones and tablet computers have made it possible for a wide range of applications to run in such an environment. In this paper, we propose the adoption of data stream classification techniques for PDM. Evident by a thorough experimental study, it has been proved that running heterogeneous/different, or homogeneous/similar data stream classification techniques over vertically partitioned data (data partitioned according to the feature space) results in comparable performance to batch and centralised learning techniques.

Sequence-specific reconstruction from fragmentary databases using seed sequences: implementation and validation on SAGE, proteome and generic sequencing data

Motivation: DNA assembly programs classically perform an all-against-all comparison of reads to identify overlaps, followed by a multiple sequence alignment and generation of a consensus sequence. If the aim is to assemble a particular segment, instead of a whole genome or transcriptome, a target-specific assembly is a more sensible approach. GenSeed is a Perl program that implements a seed-driven recursive assembly consisting of cycles comprising a similarity search, read selection and assembly. The iterative process results in a progressive extension of the original seed sequence. GenSeed was tested and validated on many applications, including the reconstruction of nuclear genes or segments, full-length transcripts, and extrachromosomal genomes. The robustness of the method was confirmed through the use of a variety of DNA and protein seeds, including short sequences derived from SAGE and proteome projects.

Slicing the metric space to provide quick indexing of complex data in the main memory

Searching in a dataset for elements that are similar to a given query element is a core problem in applications that manage complex data, and has been aided by metric access methods (MAMs). A growing number of applications require indices that must be built faster and repeatedly, also providing faster response for similarity queries. The increase in the main memory capacity and its lowering costs also motivate using memory-based MAMs. In this paper. we propose the Onion-tree, a new and robust dynamic memory-based MAM that slices the metric space into disjoint subspaces to provide quick indexing of complex data. It introduces three major characteristics: (i) a partitioning method that controls the number of disjoint subspaces generated at each node; (ii) a replacement technique that can change the leaf node pivots in insertion operations; and (iii) range and k-NN extended query algorithms to support the new partitioning method, including a new visit order of the subspaces in k-NN queries. Performance tests with both real-world and synthetic datasets showed that the Onion-tree is very compact. Comparisons of the Onion-tree with the MM-tree and a memory-based version of the Slim-tree showed that the Onion-tree was always faster to build the index. The experiments also showed that the Onion-tree significantly improved range and k-NN query processing performance and was the most efficient MAM, followed by the MM-tree, which in turn outperformed the Slim-tree in almost all the tests. (C) 2010 Elsevier B.V. All rights reserved.

A Framework for Exploring Multidimensional Data with 3D Projections

Visualization of high-dimensional data requires a mapping to a visual space. Whenever the goal is to preserve similarity relations a frequent strategy is to use 2D projections, which afford intuitive interactive exploration, e. g., by users locating and selecting groups and gradually drilling down to individual objects. In this paper, we propose a framework for projecting high-dimensional data to 3D visual spaces, based on a generalization of the Least-Square Projection (LSP). We compare projections to 2D and 3D visual spaces both quantitatively and through a user study considering certain exploration tasks. The quantitative analysis confirms that 3D projections outperform 2D projections in terms of precision. The user study indicates that certain tasks can be more reliably and confidently answered with 3D projections. Nonetheless, as 3D projections are displayed on 2D screens, interaction is more difficult. Therefore, we incorporate suitable interaction functionalities into a framework that supports 3D transformations, predefined optimal 2D views, coordinated 2D and 3D views, and hierarchical 3D cluster definition and exploration. For visually encoding data clusters in a 3D setup, we employ color coding of projected data points as well as four types of surface renderings. A second user study evaluates the suitability of these visual encodings. Several examples illustrate the framework`s applicability for both visual exploration of multidimensional abstract (non-spatial) data as well as the feature space of multi-variate spatial data.

An incremental space to visualize dynamic data sets

In Information Visualization, adding and removing data elements can strongly impact the underlying visual space. We have developed an inherently incremental technique (incBoard) that maintains a coherent disposition of elements from a dynamic multidimensional data set on a 2D grid as the set changes. Here, we introduce a novel layout that uses pairwise similarity from grid neighbors, as defined in incBoard, to reposition elements on the visual space, free from constraints imposed by the grid. The board continues to be updated and can be displayed alongside the new space. As similar items are placed together, while dissimilar neighbors are moved apart, it supports users in the identification of clusters and subsets of related elements. Densely populated areas identified in the incSpace can be efficiently explored with the corresponding incBoard visualization, which is not susceptible to occlusion. The solution remains inherently incremental and maintains a coherent disposition of elements, even for fully renewed sets. The algorithm considers relative positions for the initial placement of elements, and raw dissimilarity to fine tune the visualization. It has low computational cost, with complexity depending only on the size of the currently viewed subset, V. Thus, a data set of size N can be sequentially displayed in O(N) time, reaching O(N (2)) only if the complete set is simultaneously displayed.