Panel 1

Call for Contributions

Download the PDF version of our Call for Papers (A4 format)

Download the PDF version of our Call for Papers (letter format)

ICDL-EpiRob is a unique conference gathering researchers from computer science, robotics, psychology and developmental studies to share knowledge and research on how humans and animals develop sensing, reasoning and actions. This includes taking advantage of interaction with social and physical environments and how cognitive and developmental capabilities can be transferred to computing systems and robotics. This approach goes hand in hand with the goals of both understanding human and animal development and applying this knowledge to improve future intelligent technology, including for robots that will be in close interaction with humans.

In this ninth edition of the conference we invite submissions that explore, extend, and consolidate the interdisciplinary boundaries of this exciting research field.


Topics of interest include, but are not limited to:

  • principles and theories of development and learning;
  • development of skills in biological systems and robots;
  • nature vs nurture, developmental stages;
  • models on the contributions of interaction to learning;
  • non-verbal and multi-modal interaction;
  • models on active learning;
  • architectures for lifelong learning;
  • emergence of body and affordance perception;
  • analysis and modelling of human motion and state;
  • models for prediction, planning and problem solving;
  • models of human-human and human-robot interaction;
  • emergence of verbal and non-verbal communication;
  • epistemological foundations and philosophical issues;
  • robot prototyping of human and animal skills;
  • ethics and trust in computational intelligence and robotics;
  • social learning in humans, animals, and robots.


Accepted and presented full six-page paper submissions will be included in the conference proceedings published by IEEE Xplore.

Submit here!

Regular-length papers can also be submitted to the “MODELbot” paper challenge, to be considered for the MODELbot challenge award, see details here.

The authors of the best accepted papers will be invited to submit an extended version to be reviewed for inclusion in a “2019 ICDL-EpiRob conference” special issue of IEEE Transactions on Cognitive and Developmental Systems (TCDS).

Call for Tutorials / Workshops (Closed)

We invite experts in different areas to organise either a tutorial or a workshop to be held on the first day of the conference. Tutorials will provide insights into specific topics through hands-on training and interactive experiences. Workshops will extend to a half or full day and could include invited speakers, discussions with a smaller community, and independent submissions.

Tutorial and workshop submission are now closed!

Important dates (tentative):

  • Tutorial/workshop submission deadline: October 30, 2018
  • Paper submission deadline: February 22, 2019, 4 March 2019 (extended and final deadline)
  • Author notification: May 7, 2019
  • Camera ready due: June 1, 2019
  • Conference: August 19–22nd 2019

Go to the submission page.

Organizing committee:

  • General chairs: Jim Torresen, Kerstin Dautenhahn
  • Program chairs: Kai Olav Ellefsen, Katharina J. Rohlfing
  • Finance / Website Chairs: Kyrre Glette, Charles Martin
  • Publicity chairs: Ryo Kurazume, Bruno Castro da Silva, Kazi Shah Nawaz Ripon
  • Bridge chair: Tetsuya Ogata, Emre UgurLocal chairs: Bruno Laeng, Tor Endestad

More information about the committee.

Panel 2

Venue & Travel

ICDL-EPIROB 2019 will be held in the heart of Oslo, Norway at Ingeniørenes hus møtesenter (Engineer’s House Conference Centre). See some some general travel information in the pdf file here.

Getting to Oslo

There are two airports surrounding Oslo with Oslo airport Gardermoen (OSL) being the largest airport, where most airlines fly to. You will effectively get to downtown by taking train for slightly more than 20 minutes.

The other (and smaller) Torp Sandefjord airport (lufthavn, TRF) is located on the south-west side of the city with some international flights (low-cost airlines). Approximately 1 hour and 35–40 minutes by bus from downtown Oslo. You save a little bit on buying return bus ticket (NOK 488 regular, NOK 388 student).

Travel to Oslo from the main airport

The most effective and convenient way of getting from the main airport in Oslo (Gardermoen/OSL) is normally the Airport express train in the direction of Drammen: It runs four times per hour and cost NOK 196 one way (students half of that price, nothing to save on return tickets). However, the regular train by VY is much cheaper for regular travellers (NOK 105 regular/student price) and takes only a 2-3 more minutes but runs only 2-3 times per hour.

If staying at or near the conference venue, you should take the train to the “Nationaltheatret” station (National Theatre in English, follows after Oslo Central Station). It takes 26 minutes and arrives very close to the conference venue (Ingeniørenes Hus).

Please note that in between the Airport express train departures to Drammen, trains ending at Oslo S also run which is less convenient regarding the hotel location. Most of the regular trains by VY stops at the “Nationaltheatret” station.

Be aware that taking a taxi from the airport can be very expensive (!), especially with evening/night rates, and will also take much longer than the airport train.

About Oslo

Oslo is the centre of Norway as the seat of government but also as the economic and cultural capital. While Oslo is a small city (population of less than 1M in the urban area), it is home to a rich history, an active technology research and startup environment, vibrant theatre and music cultural events, and easy access to Norway’s beautiful natural environment.

More information about Oslo is available at the Visit Oslo web page. We look forward to offering a social program in connection with the conference that will allow attendees to experience a taste of Oslo’s culture, architecture, nature and cuisine.


Although Norway is famous for its winter weather, ICDL-EPIROB will take place in August (late summer) when the average daily high temperature is likely to be between 15-25 degrees celsius. However, weather can be anything from sunshine to cold and rainy. So be prepared. In Norway, there is no bad weather, there might only be bad clothing.

About Ingeniørenes hus

Ingeniørenes hus møtesenter located in central Oslo within an easy walk of National Theatre metro and train station, numerous accomodation options, and the exciting Aker Brygge pier, home of stylish museums and restaurants.

Image appears courtesy of VISITOSLO/Tord Baklund


Remember to consider early if you need to apply for a visa. See a list of countries not needing visa here. Don’t hesitate to get in touch with us (General Chair, Jim Torresen ) to get an invitation letter from the organizing committee. Please provide affiliation information and documentation of an accepted paper or paid registration fee when requesting a visa letter.


There are a number of hotels close to the conference venue. We recommend to book early and to compare prices on a few online booking websites (e.g.,, expedia, hrs, as prices can vary. Many tourists visit Norway at the time of the conference so booking early is expected to give the best rates. Please also check tripadvisor etc for user experiences with the different hotels.

We have negotiated special rates for ICDL-Epirob 2019 attendees at the following hotels in walking distance from the venue (limited number of rooms in each hotel):

Thon Hotel Europa (NOK 1067 for single room). Booking link. Valid until 21 July.

Thon Hotel Cecil (NOK 1317 for single room). Booking link. Valid until 1 July.

Thon Hotel Slottsparken (NOK 1113 for single room). Booking link.

Please note that we can not guarantee that these rates are lower than what you can found through other booking sites.

For budget accommodation/hostels, there are a number of options:



Power and Connectors in Norway

Norway uses the Europlug Type C & F (Schuko) connectors and 230V, 50Hz mains power, so you might need a travel adapter if coming from outside of Europe.

Panel 3


Prof Dr Stefanie Höhl


Getting attuned to others: Interpersonal synchrony and coordination in early human development

Caregiver-child interactions are characterized by interpersonal rhythms of different scales, from nursery rhymes and affective touch to daily routines. These rhythms make the environment more predictable for young children and enable interpersonal behavioral and physiological synchrony and attunement between caregiver and child. By using simultaneous measures of brain activities from caregiver and child, dual-EEG and dual-fNIRS, we can unravel the neural underpinnings of early interactional dynamics and their rhythmicity. I will present our research addressing factors critical to the establishment of caregiver-child synchrony, such as eye contact and interaction quality, especially behavioral reciprocity and contingency. I will also discuss some of the potential functions of interpersonal neural synchrony in early development, from social learning to effective cooperation and communication.

Video recording of the keynote talk.


Stefanie Höhl is Professor of Developmental Psychology at the University of Vienna. Her areas of research interest are neural and behavioral synchrony in social interactions (EEG/ fNIRS-Hyperscanning); functionality of neural rhythms in early development; social learning and communication across development; and development of face, emotion, and gaze perception.

Prof Aude Billard


Towards reproducing humans’ dexterity and reactivity

Our homes, offices and urban surroundings are carefully built to be inhabited by us, humans. Tools and furniture are designed to be easily manipulated by the human hand. Floors and stairs are modeled for human-sized legs. For robots to work seamlessly in our environments they should have bodies that resemble in shape, size and strength to the human body, and use these with the same dexterity and reactivity.

This talk will provide an overview of techniques developed in our group to enable robust, fast and flexible manipulation. Learning is guided by human demonstrations. Robust manipulation is achieved through sampling over distributions of feasible grasps. Smooth exploration leverages on complete tactile sensing coverage and learned variable impedance strategies. Bi-manual coordination offers ways to exploit the entire robot’s workspace. Imprecise positioning and sensing is overcome using active compliant strategies, similar to that displayed by humans when facing situations with high uncertainty. The talk will also present examples in which robots can learn to manipulate objects that deforms as a result of being manipulated, such as cutting fruits and vegetables.

Video recording of the keynote talk.


Aude Billard is Professor in the School of Engineering, École Polytechnique Fédérale de Lausanne. Her research interests span the control and design of robotic systems meant to interact with humans. To this goal, she pursue research in three complementary areas: a) the development of control systems for teaching robots through human demonstration; b) the study of the neural and cognitive processes underpinning imitation learning in humans; c) the design of user-friendly human-computer interfaces to facilitate human-robot interaction. She also conducts research on societal aspects of the use of robotics with application to diagnosis and therapy of children with autism. Her expertise lies in robot control, signal processing and machine learning, areas that are fundamental to her research and teaching.

Prof Hod Lipson


Automating discovery: From cognitive robotics to particle physics

Can machines discover scientific laws automatically? Despite the prevalence of big data, the process of distilling data into scientific laws has resisted automation. Particularly challenging are situations with small amounts of data that is difficult or expensive to collect. This talk will outline a series of recent research projects, starting with self-reflecting robotic systems, and ending with machines that can formulate hypotheses, design experiments, and interpret the results, to discover new scientific laws. We will see examples from psychology to cosmology, from classical physics to modern physics, from big science to small science.

Video recording of the keynote talk.


Hod Lipsonis a professor of Engineering at Columbia University in New York, and a co-author of the award winning book “Fabricated: The New World of 3D printing”, and “Driverless: Intelligent cars and the road ahead”. His work on self-aware and self-replicating robots challenges conventional views of robotics, and his TED talk on self-aware machines is one of the most viewed presentations on AI. Lipson directs the Creative Machines Lab, which pioneers new ways to make machines that create, and machines that are creative. For more information visit

Prof Michael J. Frank


Clustering and generalization of abstract structures in human reinforcement learning.
Humans are remarkably adept at generalizing knowledge between experiences in a way that can be difficult for computers.  Previous computational models and data suggest that rather than learning about each individual context, humans build latent abstract structures and learn to link these structures to arbitrary contexts, facilitating generalization. In these models, task structures that are more popular across contexts are more likely to be revisited in new contexts. However, these models predict that structures are either re-used as a whole or created from scratch, prohibiting the ability to generalize constituent parts of learned structures. This contrasts with ecological settings, where some aspects of task structure, such as the transition function, will be shared between context separately from other aspects, such as the reward function. Here, we develop a novel non-parametric Bayesian agent that forms independent latent clusters for transition and reward functions that may have different popularity across contexts. We compare this agent to an agent that jointly clusters both across a range of task domains. We show that relative performance of the two agents depends on the statistics of the task domain, including the mutual information between transition and reward functions in the environment, and the stochasticity of the observations. We formalize our analysis through an information theoretic account and develop a meta learning agent that can dynamically arbitrate between strategies across task domains, and which best fits data from human learning and generalization experiments. We argue that this provides a first step in allowing for compositional structures in reinforcement learners, which should be provide a better model of human learning and additional flexibility for artificial agents.


Michael J. Frank, PhD is Edgar L Marston Professor of Cognitive, Linguistic & Psychological Sciences and Psychiatry and Human Behavior and is affiliated with the Carney Institute for Brain Science. He directs the Brown Initiative for Computation in Brain and Mind and the Laboratory for Neural Computation. He received his PhD in Neuroscience and Psychology in 2004 at the University of Colorado, following undergraduate and master’s degrees in electrical engineering and biomedicine (Queen’s University (Canada) and University of Colorado). 

Dr. Frank’s work focuses primarily on theoretical models of frontostriatal circuits and their modulation by dopamine, especially in terms of their cognitive functions and implications for neurological and psychiatric disorders. The models are tested and refined with experiments involving pharmacological manipulation, deep brain stimulation, EEG, fMRI and genetics. Honors include Kavli Fellow (2016), the Cognitive Neuroscience Society Young Investigator Award (2011), the Janet T Spence Award for early career transformative contributions (Association for Psychological Science, 2010) and the DG Marquis award for best paper published in Behavioral Neuroscience (2006). Dr Frank is a senior editor for eLife, associate editors for Behavioral Neuroscience and the Journal of Neuroscience, and member of Faculty of 1000 (Theoretical Neuroscience section).

Panel 4

MODELbot Challenge

The MODELBot Challenge: MOdelling DEvelopment and Learning with robots and computers

The field of developmental robotics forms a bridge between two research communities: those who study learning and development in humans and animals, and those who study comparable processes in artificial systems. MODELbot–MOdelling DEvelopment and Learning with robots and computers–Challenge papers are expected to help strengthen this bridge, by establishing a strong link between developmental studies on the one hand and robotic and /or computational modeling on the other hand. We invite both models inspired by / studying human and animal development and learning.

To reflect the interdisciplinary focus of ICDL-Epirob, we are also this year welcoming submissions demonstrating the use of robotic and computational techniques for supporting human learning (e.g. studies in child-robot interaction). Submissions will be judged by the following criteria:

  • How well is the computational model (e.g. an artificial system, which can be a robot or a software agent) grounded in recent studies of human or animal development and/or learning?
  • The extent of the novel insights or explanations generated by the model or interaction studies, and importantly whether the model makes interesting and testable predictions.
  • How well presented and analyzed are the robotic/computational model and experimental findings, and how well written is the paper.

“MODELbot Challenge” papers are submitted through the regular submission portal, by selecting submit-link next to the MODELBot Challenge.

We link to a few previous submissions to the competition below, for inspiration. However, notice that the scope this year has been extended, and therefore a wider range of papers are welcome for submission.