[Comp-neuro] Intrinsic motivations: announcement of several works
Gianluca Baldassarre <firstname.lastname@example.org>
gianluca.baldassarre at gmail.com
Mon Feb 16 16:19:16 CET 2015
I announce the publication of various collections of papers and activities
on intrinsic motivations (IMs). Please find below a list of brief
explanations and links to them.
Further below I also briefly introduce intrinsic motivations for those who
are not familiar with them.
*** Collections of papers and activities on IMs ***
Lately I co-edited a special issue (`Topic') of Frontiers.
The articles of the Topic were published in part with `Frontiers in
Psychology - Cognitive Science' (19 articles):
and in part with `Frontiers in Neurorobotics' (6 articles):
All articles of the Topic, including an editorial that briefly introduces
IMs and summarises the contributions, can be downloaded with one click from
this web page as a `Frontiers eBook' (1 pdf file of 45 MB):
Before the Topic, we edited a Springer book collecting other 17 papers on
`Baldassarre G., Mirolli M. (eds) (2013). Intrinsically motivated learning
in natural and artificial systems. Berlin: Springer.'
You can see the titles, abstracts, authors of the book chapters, and other
information on the book, here:
We also edited a second Springer book collecting further 14 papers:
`Baldassarre G., Mirolli M. (eds) (2013). Computational and Robotic Models
of the Hierarchical Organisation of Behaviour. Berlin: Springer-Verlag.'
Hierarchical architectures represent a key problem for IMs as the knowledge
and skills accumulated under the drive of IMs has to be stored in suitable
hierarchical systems. You can see the titles, abstracts and authors of the
book chapters, and other information on the book, here:
The three collections above were prompted and partially funded by a 4-year
EU-funded project, now terminated, called `IM-CLeVeR -- Intrinsically
Motivated Cumulative Learning Robots'. The project was funded by the
European Commission under the 7th Framework Programme (FP7/2007-2013), ICT
Challenge 2 `Cognitive Systems and Robotics' (grant agreement no.
ICT-IP-231722). The web-site of the project is here:
In the project web-site you can find information on the project activities
and other material on IMs, for example a summary of main results and
publications of the project:
a final digest of the project insights:
and a `tool-box' of information for those who are interested in IMs:
I hope these works contribute to attract an increasing attention of...
clever and strongly intrinsically-motivated researchers from different
disciplines on intrinsic motivations, a great topic of investigation so
important for humans and for technology.
*** Brief introduction to intrinsic motivations (IMs) ***
Intrinsic motivations are related to things such as curiosity, the interest
for novel stimuli and surprising events, and the motivation to learn new
behaviours (think about children at play). It has been proposed that the
adaptive value of IMs is to motivate and guide the cumulative acquisition
of knowledge and skills that can be later used (e.g., in adulthood) to
accomplish goals that enhance biological fitness. IMs continue to operate
also during adulthood and indeed in humans they underlie life-long learning
and typically human activities such as art and scientiﬁc discovery. IMs are
also at the core of processes that strongly aﬀect human well-being, such as
the sense of competence, self-determination, and self-esteem.
Recent neuroscientiﬁc research is starting to uncover the basic brain
mechanisms underlying IMs, although under research agendas not directly
addressing IMs. For example, IM mechanisms have been related to learning
signals based on dopamine, novelty detection in hippocampus, and prediction
errors in various parts of brain.
In the last decade, IMs have also been introduced in machine learning and
autonomous robotics as a means for developing artiﬁcial systems that can
autonomously learn several diﬀerent skills in an open-ended, cumulative
fashion. The idea is that intelligent machines and robots could
autonomously acquire skills and knowledge under the guidance of IMs, and
later exploit such knowledge and skills to accomplish the tasks that are
useful for the user in more eﬃcient and faster ways than if they had to
acquire them from scratch. This possibility would especially enhance the
utility of intelligent artiﬁcial systems when operating in real-life
environments posing challenges that cannot be foreseen at design time.
Although the possible functions and mechanisms of IMs are still debated and
seen differently by the different research communities involved, and
although the recognition of IMs as a research field per se is only
partially accomplished, their link with some of the most sophisticated
aspects of human cognition (curiosity, art, science, well-being), and their
potential for applications make them an interesting and important research
field that deserves additional research efforts.
Gianluca Baldassarre, Ph.D.,
Laboratory of Computational Embodied Neuroscience,
Istituto di Scienze e Tecnologie della Cognizione,
Consiglio Nazionale delle Ricerche (LOCEN-ISTC-CNR),
Via San Martino della Battaglia 44, I-00185 Roma, Italy
E-mail: gianluca.baldassarre at istc.cnr.it
Tel: +39 06 44 595 231
Fax: +39 06 44 595 243
Learn from the past, live in(tensely) the present, dream for the future
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Comp-neuro