arXiv subjects

Giovanni Granato

Publications and source records attributed to Giovanni Granato.

5 recordsLinked to original sources

arXiv2023

A Neurocomputational Account of Flexible Goal-directed Cognition and Consciousness: The Goal-Aligning Representation Internal Manipulation Theory (GARIM)

Goal-directed manipulation of representations is a key element of human flexible behaviour, while consciousness is often related to several aspects of higher-order cognition and human flexibility. Currently these two phenomena are only partially integrated (e.g., see Neurorepresentationalism) and this (a) limits our understanding of neuro-computational processes that lead conscious states to produce flexible goal-directed behaviours, (b) prevents a computational formalisation of conscious goal-directed manipulations of representations occurring in the brain, and (c) inhibits the exploitation of this knowledge for modelling and technological purposes. Addressing these issues, here we extend our `three-component theory of flexible cognition' by proposing the `Goal-Aligning Representations Internal Manipulation' (GARIM) theory of conscious and flexible goal-directed cognition. The central idea of the theory is that conscious states support the active manipulation of goal-relevant internal representations (e.g., of world states, objects, and action sequences) to make them more aligned with the pursued goals. This leads to the generation of the knowledge which is necessary to face novel situations/goals, thus increasing the flexibility of goal-directed behaviours. The GARIM theory integrates key aspects of the main theories of consciousness into the functional neuro-computational framework of goal-directed behaviour. Moreover, it takes into account the subjective sensation of agency that accompanies conscious goal-directed processes (`GARIM agency'). The proposal has also implications for experimental studies on consciousness and clinical aspects of conscious goal-directed behaviour. Finally, the GARIM theory benefit technological fields such as autonomous robotics and machine learning (e.g., the manipulation process may describe the operations performed by systems based on transformers).

cs.AI↗

arXiv2021

A Computational Model of Representation Learning in the Brain Cortex, Integrating Unsupervised and Reinforcement Learning

A common view on the brain learning processes proposes that the three classic learning paradigms -- unsupervised, reinforcement, and supervised -- take place in respectively the cortex, the basal-ganglia, and the cerebellum. However, dopamine outbursts, usually assumed to encode reward, are not limited to the basal ganglia but also reach prefrontal, motor, and higher sensory cortices. We propose that in the cortex the same reward-based trial-and-error processes might support not only the acquisition of motor representations but also of sensory representations. In particular, reward signals might guide trial-and-error processes that mix with associative learning processes to support the acquisition of representations better serving downstream action selection. We tested the soundness of this hypothesis with a computational model that integrates unsupervised learning (Contrastive Divergence) and reinforcement learning (REINFORCE). The model was tested with a task requiring different responses to different visual images grouped in categories involving either colour, shape, or size. Results show that a balanced mix of unsupervised and reinforcement learning processes leads to the best performance. Indeed, excessive unsupervised learning tends to under-represent task-relevant features while excessive reinforcement learning tends to initially learn slowly and then to incur in local minima. These results stimulate future empirical studies on category learning directed to investigate similar effects in the extrastriate visual cortices. Moreover, they prompt further computational investigations directed to study the possible advantages of integrating unsupervised and reinforcement learning processes.

q-bio.NC↗

arXiv2012

Level-set approach for Reachability Analysis of Hybrid Systems under Lag Constraints

This study aims at characterizing a reachable set of a hybrid dynamical system with a lag constraint in the switch control. The setting does not consider any controllability assumptions and uses a level-set approach. The approach consists in the introduction of on adequate hybrid optimal control problem with lag constraints on the switch control whose value function allows a characterization of the reachable set. The value function is in turn characterized by a system of quasi-variational inequalities (SQVI). We prove a comparison principle for the SQVI which shows uniqueness of its solution. A class of numerical finite differences schemes for solving the system of inequalities is proposed and the convergence of the numerical solution towards the value function is studied using the comparison principle. Some numerical examples illustrating the method are presented. Our study is motivated by an industrial application, namely, that of range extender electric vehicles. This class of electric vehicles uses an additional module -- the range extender -- as an extra source of energy in addition to its main source -- a high voltage battery. The reachability study of this system is used to establish the maximum range of a simple vehicle model.

math.OC↗

arXiv2012

Reachability of Delayed Hybrid Systems Using Level-set Methods

This study proposes an algorithm to synthesize controllers for the power management on board hybrid vehicles that allows the vehicle to reach its maximum range along a given route. The algorithm stems from a level-set approach that computes the reachable set of the system, i.e., the collection of states reachable from a certain initial condition via the computation of the value function of an optimal control problem. The discrete-time vehicle model is one of a particular class of hybrid vehicles, namely, range extender electric vehicles (REEV). This kind of hybridization departures from a full electric vehicle that has an additional module -- the range extender (RE) -- as an extra energy source in addition to its main energy source -- a high voltage battery. As an important feature, our model allows for the switching on and off of the range extender and includes a decision lag constraint, i.e., imposes two consecutive switches to be separated by a positive time interval. The approach consists in the introduction of an adequate optimal control problem with lag constraints on the switch control whose value function allows a characterization of the reachable set. The value function is in turn characterized by a dynamic programming algorithm. This algorithm is implemented and some numerical examples are presented.

math.OC↗

arXiv2012

A Stochastic Dynamic Principle for Hybrid Systems with Execution Delay and Decision Lags

This work presents a stochastic dynamic programming (SDP) algorithm that aims at minimizing an economic criteria based on the total energy consumption of a range extender electric vehicle (REEV). This algorithm integrates information from the REEV's navigation system in order to obtain some information about future expected vehicle speed. The model of the vehicle's energetic system, which consists of a high-voltage (HV) battery, the main energy source, and an internal combustion engine (ICE), working as an auxiliary energy source), is written as a hybrid dynamical system and the associated optimization problem in the hybrid optimal control framework. The hybrid optimal control problem includes two important physical constraints on the ICE, namely, an activation delay and a decision lag. Three methods for the inclusion of such physical constraints are studied. After introducing the SDP algorithm formulation we comment on numerical results of the stochastic algorithm and its deterministic counterpart.

math.OC↗