STATE OF THE ART IN BRAIN ARCHITECTURES. DAC – DISTRIBUTED ADAPTIVE CONTROL.
Riccardo Zucca, Jordi Ysard Puigbó
This tutorial introduces the Distributed Adaptive Control (DAC), a theory of the design principles underlying the Mind, Brain, Body Nexus (MBBN) that has been developed over the last 20 years. DAC assumes that the brain maintains stability between an embodied agent, its internal state and its environment through action. It postulates that in order to act, or know how, the brain has to answer 5 fundamental questions: who, why, what, where, when. Thus, the function of the brain is to continuously solve the so-called H5W problem with ‘H’ standing for the ‘How’ an agent acts in the world. The DAC theory is expressed as a neural-based architecture implemented in robots and organized in two complementary structures: layers and columns. The organizational layers are called: reactive, adaptive and contextual, and its columnar organization defines the processing of states of the world, the self and the generation of action. After an overview of the key elements of DAC, the mapping of its key assumptions towards the invertebrate and mammalian brain is described. The general overview of DAC’s explanation of MBBN is combined with examples of application scenarios in which DAC has been validated, including mobile and humanoid robots, neuro-rehabilitation and the large scale interactive spaces. In this tutorial we will provide the elements necessary to implement an autonomous control system based on the DAC architecture and we will explore how the different layers of DAC contribute to solve a foraging task.
BRAIN X3 – BRAIN FUNCTIONAL NETWORK CONNECTIVITY.
Xerxes Arsiwalla, Toni Gurgui. IBEC, Barecelona
With this tutorial, we address multi-modal data integration in human brain imaging using BrainX3, an interactive platform for 3D visualization, analysis, and simulation of human neuroimaging data. We will focus on volumetric MRI data, DTI/DSI tractographic data, intracranial epilepsy data, and semantic corpora from available text databases. BrainX3 provides a tool to organize and visualize data and extract meaningful insights about brain structures and functional networks. BrainX3 is a useful tool for scientific exploration and in the clinic to investigate complex neuro-pathologies and for neuro-surgery. BrainX3 is developed by SPECS research group to advance knowledge in mind-brain and behaviour
DAC – FORAGING BEHAVIOR (simulations library) AND HUMAN ROBOT INTERACTION.
Vicky Vouloutsi, Sean O’Sullivan. IBEC, Barcelona.
DAC-X: PREDICTION, ACTION SELECTION, MOTOR CONTROL.
Diogo Pata, Adrian Fernandez Amil. IBEC, Barcelona.
Animals successfully forage within new environments by learning, simulating and adapting to their surroundings. The functions behind such goal-oriented behavior can be decomposed into 5 top-level objectives: ‘how’, ‘why’, ‘what’, ‘where’, ‘when’ (H4W). The paradigms of classical and operant conditioning describe some of the behavioral aspects found in foraging. However, it remains unclear how the organization of their underlying neural principles account for these complex behaviors. We address this problem from the perspective of the Distributed Adaptive Control theory of mind and brain (DAC) that interprets these two paradigms as expressing properties of core functional subsystems of a layered architecture. In particular, we propose DAC-X, a novel cognitive architecture that unifies the theoretical principles of DAC with biologically constrained computational models of several areas of the mammalian brain …
DEEP LEARNING TECNIQUES FOR VISUAL PERCEPTION.
Georgios Th. Papadopoulos. Centre for Research and Technology Hellas, Thessaloniki, Greece
This tutorial will initially demonstrate the fundamental principles of the deep learning paradigm and its relation to the overall machine learning field. Subsequently, a detailed description of the theory and the building blocks of the most commonly met types of neural networks (namely convolutional and recurrent networks) will be provided. Then, insights on the evolution of deep learning architectures (with emphasis on key milestone works) will be given. The tutorial will conclude with the presentation of indicative applications of deep learning schemes in different visual perception tasks.
SITUATED AND EMBODIED PROCESSES: A FRAMEWORK TO APPROACH LIVE INTERACTIVE COMPOSITION. Jônatas Manzolli. NICS, University of Campinas (UNICAMP), Brazil
This tutorial focus on the convergences between computer-based systems developed around live interactive composition and situated music with processes that have emerged from models in computational neuroscience and music information retrieval. Firstly, I’ll describe a conceptual point of view according to which a theory of mind can be applied to the development of interactive computation models that produced sound material in real-time. Then, new compositional trends such as the use of extended instrumental techniques are associated with the development of new interfaces for music expression, methods for sound synthesis using digital instruments, and devices for human-machine interaction. These aspects, once interconnected, might lead to new research approaches in analyzing the nature of the sound phenomena and the various possibilities of extracting information from sound pressure waves via spectral analysis. This emphasis on sound as such, combined with new technologies for its synthesis and combination in complex compositions, let creative approaches emerge that move away from composing a particular musical piece to designing a potential space of musical expression, where the particular will be defined through interaction between the musical system and its environment. It can then be speculated that innovative comparative music and language research could follow, striving for more general and process-oriented theories of mind, brain, and behavior. This, in turn, enables comparative research on action, language, and music. Such a development would be based on the dynamics of interaction and embodiment. Overall, it is an action-oriented approach to cognition, perception, and emotion in the spirit of motor theories of cognition. These direct research on music cognition and computational modeling away from score only approaches and towards studying musical behavior and the music-making of agents. To illustrate this perspective, I present a set of live interactive compositions that experiment with such situated aesthetics.