Autonomous Reinforcement Learning of Multiple Interrelated Tasks

Santucci, Vieri Giuliano; Baldassarre, Gianluca; Cartoni, Emilio

doi:10.1109/devlrn.2019.8850713

Cited by 18 publications

(15 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the computational literature, IMs have been implemented to foster different types of autonomous behaviors such as state-space exploration (Bellemare et al 2016;Romero et al 2020;Schillaci et al 2020), knowledge gathering (Schmidhuber 2010), learning repertoire of skills (Singh et al 2004;Oudeyer et al 2013), affordance exploration and exploitation (Hart and Grupen 2013;Baldassarre et al 2019;Manoury et al 2019). Furthermore, and closely related to the topic discussed in this article, IMs have been used to allow embodied artificial agents to autonomously discover and select their own goals (Baranes and Oudeyer 2013;Santucci et al 2016Santucci et al , 2019Blaes et al 2020). Instead of having specific tasks assigned to them, artificial agents are left free to explore the environment according to criteria such as novelty, unexpected events, or the improvement of their ability to achieve autonomously selected goals.…”

Section: Machine Learning Approaches For Autonomous Artificial Agentsmentioning

confidence: 99%

Interdependence as the key for an ethical artificial autonomy

Pianca

Santucci

2022

AI & Soc

Self Cite

View full text Add to dashboard Cite

Currently, the autonomy of artificial systems, robotic systems in particular, is certainly one of the most debated issues, both from the perspective of technological development and its social impact and ethical repercussions. While theoretical considerations often focus on scenarios far beyond what can be concretely hypothesized from the current state of the art, the term autonomy is still used in a vague or too general way. This reduces the possibilities of a punctual analysis of such an important issue, thus leading to often polarized positions (naive optimism or unfounded defeatism). The intent of this paper is to clarify what is meant by artificial autonomy, and what are the prerequisites that can allow the attribution of this characteristic to a robotic system. Starting from some concrete examples, we will try to indicate a way towards artificial autonomy that can hold together the advantages of developing adaptive and versatile systems with the management of the inevitable problems that this technology poses both from the viewpoint of safety and ethics. Our proposal is that a real artificial autonomy, especially if expressed in the social context, can only be achieved through interdependence with other social actors (human and otherwise), through continuous exchanges and interactions which, while allowing robots to explore the environment, guarantee the emergence of shared practices, behaviors, and ethical principles, which otherwise could not be imposed with a top-down approach, if not at the price of giving up the same artificial autonomy.

show abstract

Section: Machine Learning Approaches For Autonomous Artificial Agentsmentioning

confidence: 99%

Interdependence as the key for an ethical artificial autonomy

Pianca

Santucci

2022

AI & Soc

Self Cite

View full text Add to dashboard Cite

show abstract

“…In a very similar approach, Santucci et al [21] demonstrated the performance of the open-ended Goal-Discovering Robotic Architecture for Intrinsically-Motivated Learning (GRAIL). The proposed system discovers interesting events while it interacts with the environment and sets "goals", which are later used to drive the learning in an intrinsically motivated manner.…”

Section: Related Studiesmentioning

confidence: 99%

Discovering Schema-Based Action Sequences Through Play in Situated Humanoid Robots

Kumar

Giagkos

Shaw

et al. 2022

IEEE Trans. Cogn. Dev. Syst.

View full text Add to dashboard Cite

Exercising sensorimotor and cognitive functions allows humans, including infants, to interact with the environment and objects within it. In particular, during everyday activities, infants continuously enrich their repertoire of actions, and by playing, they experimentally plan such actions in sequences to achieve desired goals. The latter, reflected as perceptual target states, are built on previously acquired experiences shaped by infants to predict their actions. Imitating this, in developmental robotics, we seek methods that allow autonomous embodied agents with no prior knowledge to acquire information about the environment. Like infants, robots that actively explore the surroundings and manipulate proximate objects are capable of learning. Their understanding of the environment develops through the discovery of actions and their association with the resulting perceptions in the world. We extend the development of Dev-PSchema, a schema-based, open-ended learning system, and examine the infant-like discovery process of new generalised skills while engaging with objects in free-play using an iCub robot. Our experiments demonstrate the capability of Dev-PSchema to utilise the newly discovered skills to solve user-defined goals beyond its past experiences. The robot can generate and evaluate sequences of interdependent high-level actions to form potential solutions and ultimately solve complex problems towards tool-use.

show abstract

“…The SEEKING drive perspective connects with another enormous research fields: the one about intrinsic vs. extrinsic motivation (e.g., Santucci et al, 2019 ), which – in turn – relates to the information-SEEKING research topic (e.g., Horan et al, 2019 ). Although there are many interpretations about the roles of the ML DA system (e.g., Bromberg-Martin et al, 2010 ; FitzGerald et al, 2015 ; Schultz, 2015 ), the SEEKING drive theory has the advantage of subsuming all other functions within a single basic psycho-behavioral disposition (see Alcaro et al, 2007 for a review).…”

Section: The ML Da-seeking Systemmentioning

confidence: 99%

The SEEKING Drive and Its Fixation: A Neuro-Psycho-Evolutionary Approach to the Pathology of Addiction

Alcaro

Brennan

Conversi

2021

Front. Hum. Neurosci.

View full text Add to dashboard Cite

Neuro-ethological studies conducted by Panksepp and his colleagues have provided an understanding of how the activity of the mesolimbic dopaminergic (ML DA) system leads to the emotional disposition to SEEK/Explore, which is involved in all appetitive motivated behavior and mental activity. In pathological addiction phenomena, this emotional disposition “fixes” itself on certain obsessive-compulsive habits, losing its versatility and its natural predisposition to spontaneous and unconditioned activation. Overall, the result is a consistent disinterest in everything that is not the object of addiction. From a neuro-psycho-evolutionary point of view, the predisposition to develop addictive behavior can be attributed to a loss of “functional autonomy” of the SEEKING/Explorative disposition. Indeed, as shown by animal and human studies, the tendency to be conditioned by situations and contexts that provide an immediate reward can be closely related to a deficit in the tonic endogenous activity of the ML DA-SEEKING system.

show abstract

Autonomous Reinforcement Learning of Multiple Interrelated Tasks

Cited by 18 publications

References 59 publications

Interdependence as the key for an ethical artificial autonomy

Interdependence as the key for an ethical artificial autonomy

Discovering Schema-Based Action Sequences Through Play in Situated Humanoid Robots

The SEEKING Drive and Its Fixation: A Neuro-Psycho-Evolutionary Approach to the Pathology of Addiction

Contact Info

Product

Resources

About