Showing papers by "Giovanni Pezzulo published in 2014"

Consensus Paper: The Cerebellum's Role in Movement and Cognition

Abstract: While the cerebellum's role in motor function is well recognized, the nature of its concurrent role in cognitive function remains considerably less clear. The current consensus paper gathers diverse views on a variety of important roles played by the cerebellum across a range of cognitive and emotional functions. This paper considers the cerebellum in relation to neurocognitive development, language function, working memory, executive function, and the development of cerebellar internal control models and reflects upon some of the ways in which better understanding the cerebellum's status as a “supervised learning machine” can enrich our ability to understand human function and adaptation. As all contributors agree that the cerebellum plays a role in cognition, there is also an agreement that this conclusion remains highly inferential. Many conclusions about the role of the cerebellum in cognition originate from applying known information about cerebellar contributions to the coordination and quality of movement. These inferences are based on the uniformity of the cerebellum's compositional infrastructure and its apparent modular organization. There is considerable support for this view, based upon observations of patients with pathology within the cerebellum.

The why, what, where, when and how of goal-directed choice: neuronal and computational principles

Abstract: The central problems that goal-directed animals must solve are: ‘What do I need and Why, Where and When can this be obtained, and How do I get it?' or the H4W problem. Here, we elucidate the principles underlying the neuronal solutions to H4W using a combination of neurobiological and neurorobotic approaches. First, we analyse H4W from a system-level perspective by mapping its objectives onto the Distributed Adaptive Control embodied cognitive architecture which sees the generation of adaptive action in the real world as the primary task of the brain rather than optimally solving abstract problems. We next map this functional decomposition to the architecture of the rodent brain to test its consistency. Following this approach, we propose that the mammalian brain solves the H4W problem on the basis of multiple kinds of outcome predictions, integrating central representations of needs and drives (e.g. hypothalamus), valence (e.g. amygdala), world, self and task state spaces (e.g. neocortex, hippocampus and prefrontal cortex, respectively) combined with multi-modal selection (e.g. basal ganglia). In our analysis, goal-directed behaviour results from a well-structured architecture in which goals are bootstrapped on the basis of predefined needs, valence and multiple learning, memory and planning mechanisms rather than being generated by a singular computation.

Why do you fear the bogeyman? An embodied predictive coding model of perceptual inference.

Abstract: Why are we scared by nonperceptual entities such as the bogeyman, and why does the bogeyman only visit us during the night? Why does hearing a window squeaking in the night suggest to us the unlikely idea of a thief or a killer? And why is this more likely to happen after watching a horror movie? To answer these and similar questions, we need to put mind and body together again and consider the embodied nature of perceptual and cognitive inference. Predictive coding provides a general framework for perceptual inference; I propose to extend it by including interoceptive and bodily information. The resulting embodied predictive coding inference permits one to compare alternative hypotheses (e.g., is the sound I hear generated by a thief or the wind?) using the same inferential scheme as in predictive coding, but using both sensory and interoceptive information as evidence, rather than just considering sensory events. If you hear a window squeaking in the night after watching a horror movie, you may consider plausible a very unlikely hypothesis (e.g., a thief, or even the bogeyman) because it explains both what you sense (e.g., the window squeaking in the night) and how you feel (e.g., your high heart rate). The good news is that the inference that I propose is fully rational and gives minds and bodies equal dignity. The bad news is that it also gives an embodiment to the bogeyman, and a reason to fear it.

The principles of goal-directed decision-making: from neural mechanisms to computation and robotics.

Abstract: The brain has evolved to act in a complex and unpredictable world, and it must continuously select among many goals and action options. In the past 20 years, research in neuroscience and robotics has principally focused on the neural underpinnings of habitual choice, linked to stimulus–response

Decision and action planning signals in human posterior parietal cortex during delayed perceptual choices

Abstract: During simple perceptual decisions, sensorimotor neurons in monkey fronto-parietal cortex represent a decision variable that guides the transformation of sensory evidence into a motor response, supporting the view that mechanisms for decision-making are closely embedded within sensorimotor structures. Within these structures, however, decision signals can be dissociated from motor signals, thus indicating that sensorimotor neurons can play multiple and independent roles in decision-making and action selection/planning. Here we used functional magnetic resonance imaging to examine whether response-selective human brain areas encode signals for decision-making or action planning during a task requiring an arbitrary association between face pictures (male vs. female) and specific actions (saccadic eye vs. hand pointing movements). The stimuli were gradually unmasked to stretch the time necessary for decision, thus maximising the temporal separation between decision and action planning. Decision-related signals were measured in parietal and motor/premotor regions showing a preference for the planning/execution of saccadic or pointing movements. In a parietal reach region, decision-related signals were specific for the stimulus category associated with its preferred pointing response. By contrast, a saccade-selective posterior intraparietal sulcus region carried decision-related signals even when the task required a pointing response. Consistent signals were observed in the motor/premotor cortex. Whole-brain analyses indicated that, in our task, the most reliable decision signals were found in the same neural regions involved in response selection. However, decision- and action-related signals within these regions can be dissociated. Differences between the parietal reach region and posterior intraparietal sulcus plausibly depend on their functional specificity rather than on the task structure.

The Cat is on the Mat. or is it a Dog? Dynamic Competition in Perceptual Decision Making

Abstract: Recent neurobiological findings suggest that the brain solves simple perceptual decision-making tasks by means of a dynamic competition in which evidence is accumulated in favor of the alternatives. However, it is unclear if and how the same process applies in more complex, real-world tasks, such as the categorization of ambiguous visual scenes and what elements are considered as evidence in this case. Furthermore, dynamic decision models typically consider evidence accumulation as a passive process disregarding the role of active perception strategies. In this paper, we adopt the principles of dynamic competition and active vision for the realization of a biologically-motivated computational model, which we test in a visual categorization task. Moreover, our system uses predictive power of the features as the main dimension for both evidence accumulation and the guidance of active vision. Comparison of human and synthetic data in a common experimental setup suggests that the proposed model captures essential aspects of how the brain solves perceptual ambiguities in time. Our results point to the importance of the proposed principles of dynamic competition, parallel specification, and selection of multiple alternatives through prediction, as well as active guidance of perceptual strategies for perceptual decision-making and the resolution of perceptual ambiguities. These principles could apply to both the simple perceptual decision problems studied in neuroscience and the more complex ones addressed by vision research.

Topological self-organization and prediction learning support both action and lexical chains in the brain.

Abstract: A growing body of evidence in cognitive psychology and neuroscience suggests a deep interconnection between sensory-motor and language systems in the brain. Based on recent neurophysiological findings on the anatomo-functional organization of the fronto-parietal network, we present a computational model showing that language processing may have reused or co-developed organizing principles, functionality, and learning mechanisms typical of premotor circuit. The proposed model combines principles of Hebbian topological self-organization and prediction learning. Trained on sequences of either motor or linguistic units, the network develops independent neuronal chains, formed by dedicated nodes encoding only context-specific stimuli. Moreover, neurons responding to the same stimulus or class of stimuli tend to cluster together to form topologically connected areas similar to those observed in the brain cortex. Simulations support a unitary explanatory framework reconciling neurophysiological motor data with established behavioral evidence on lexical acquisition, access, and recall.

The sensorimotor and social sides of the architecture of speech.

Abstract: Speech is a complex skill to master. In addition to sophisticated phono-articulatory abilities, speech acquisition requires neuronal systems configured for vocal learning, with adaptable sensorimotor maps that couple heard speech sounds with motor programs for speech production; imitation and self-imitation mechanisms that can train the sensorimotor maps to reproduce heard speech sounds; and a "pedagogical" learning environment that supports tutor learning.

Bounded Seed-AGI

Abstract: Four principal features of autonomous control systems are left both unaddressed and unaddressable by present-day engineering methodologies: (1) The ability to operate effectively in environments that are only partially known at design time; (2) A level of generality that allows a system to re-assess and re-define the fulfillment of its mission in light of unexpected constraints or other unforeseen changes in the environment; (3) The ability to operate effectively in environments of significant complexity; and (4) The ability to degrade gracefully—how it can continue striving to achieve its main goals when resources become scarce, or in light of other expected or unexpected constraining factors that impede its progress. We describe new methodological and engineering principles for addressing these shortcomings, that we have used to design a machine that becomes increasingly better at behaving in underspecified circumstances, in a goal-directed way, on the job, by modeling itself and its environment as experience accumulates. The work provides an architectural blueprint for constructing systems with high levels of operational autonomy in underspecified circumstances, starting from only a small amount of designer-specified code—a seed. Using value-driven dynamic priority scheduling to control the parallel execution of a vast number of lines of reasoning, the system accumulates increasingly useful models of its experience, resulting in recursive self-improvement that can be autonomously sustained after the machine leaves the lab, within the boundaries imposed by its designers. A prototype system named AERA has been implemented and demonstrated to learn a complex real-world task—real-time multimodal dialogue with humans—by on-line observation. Our work presents solutions to several challenges that must be solved for achieving artificial general intelligence.

Goals reconfigure cognition by modulating predictive processes in the brain.

Abstract: I applaud Huang & Bargh's (H&B's) theory that places goals at the center of cognition, and I discuss two ingredients missing from that theory. First, I argue that the brains of organisms much simpler than those of humans are already configured for goal achievement in situated interactions. Second, I propose a mechanistic view of the "reconfiguration principle" that links the theory with current views in computational neuroscience.

Autonomous acquisition of natural language

Abstract: An important part of human intelligence is the ability to use language. Humans learn how to use language in a society of language users, which is probably the most effective way to learn a language from the ground up. Principles that might allow an artificial agents to learn language this way are not known at present. Here we present a framework which begins to address this challenge. Our auto-catalytic, endogenous, reflective architecture (AERA) supports the creation of agents that can learn natural language by observation. We present results from two experiments where our S1 agent learns human communication by observing two humans interacting in a realtime mock television interview, using gesture and situated language. Results show that S1 can learn multimodal complex language and multimodal communicative acts, using a vocabulary of 100 words with numerous sentence formats, by observing unscripted interaction between the humans, with no grammar being provided to it a priori, and only high-level information about the format of the human interaction in the form of high-level goals of the interviewer and interviewee and a small ontology. The agent learns both the pragmatics, semantics, and syntax of complex sentences spoken by the human subjects on the topic of recycling of objects such as aluminum cans, glass bottles, plastic, and wood, as well as use of manual deictic reference and anaphora.

Autonomous Acquisition of Natural Situated Communication

Abstract: An important part of human intelligence, both historically and operationally, is our ability to communicate. We learn how to communicate, and maintain our communicative skills, in a society of communicators – a highly effective way to reach and maintain proficiency in this complex skill. Principles that might allow artificial agents to learn language this way are in completely known at present – the multi-dimensional nature of socio-communicative skills are beyond every machine learning framework so far proposed. Our work begins to address the challenge of proposing a way for observation-based machine learning of natural language and communication. Our framework can learn complex communicative skills with minimal up-front knowledge. The system learns by incrementally producing predictive models of causal relationships in observed data, guided by goal-inference and reasoning using forward-inverse models. We present results from two experiments where our S1 agent learns human communication by observing two humans interacting in a realtime TV-style interview, using multimodal communicative gesture and situated language to talk about recycling of various materials and objects. S1 can learn multimodal complex language and multimodal communicative acts, a vocabulary of 100 words forming natural sentences with relatively complex sentence structure, including manual deictic reference and anaphora. S1 is seeded only with high-level information about goals of the interviewer and interviewee, and a small ontology; no grammar or other information is provided to S1 a priori. The agent learns the pragmatics, semantics, and syntax of complex utterances spoken and gestures from scratch, by observing the humans compare and contrast the cost and pollution related to recycling aluminum cans, glass bottles, newspaper, plastic, and wood. After 20 hours of observation S1 can perform an unscripted TV interview with a human, in the same style, without making mistakes.