AI and Education

The research theme on AI and education aims at applying some of the IIIA AI techniques to the field of education. It does not aim at completely automating some processes and replacing the human component, but supporting the human through mechanisms like team formation that allow for building more efficient teams, or peer-based assessments that support assessments in massive online courses. 

Contact: Carles Sierra

It is always challenging to make predictions about the impact of technology on an economic or social sector. However, all recent analysis make it clear that repetitive tasks, or those with little added value by the humans who perform them, are going to be redesigned to facilitate their automation through the use of Artificial Intelligence (AI) techniques. Banking and commerce are examples of sectors that are undergoing a profound transformation, partly enabled by AI techniques such as chat-bots or personalization systems, leading to a notable reduction in employment. On the contrary, the education sector will continue to need the human component, and permanently, since the school fulfils an essential socializing function for the development of people. This need does not mean that AI is not going to impact educational processes; it has; it does and will continue to do so.    

Today we have numerous AI applications, not necessarily developed specifically for education, but which are very useful in the education world. For example, automatic subtitling of videos, tutoring systems that interact in natural language, or the realistic conversion of text to human speech. 

AI, in its origins, was already applied to education, in particular personalized education. Adapting the contents to each student is a pedagogical imperative that is difficult for teachers to achieve when the groups are large or the economic resources dedicated to education are limited. Several research groups developed simple systems of personalized education in the 1960s. Today, these systems have reached a remarkable level of sophistication. For example, over the past 15 years, the ALEKS system ( developed in the United States has improved the performance of millions of students in mathematics. This system raises problems with an open response, analyses the answer and, thanks to a machine learning system, identifies errors and skills not acquired to explain the error to the student and recommend new problems that help to obtain the necessary skills. This type of system continues to be developed in different countries. The one created by Squirrel AI in Shanghai with more than three million students, and with a great improvement in individual performance is noteworthy ( We will no doubt see these systems more frequently over the next decade and covering areas increasingly distant from STEM, where they focus on today.

Collaborative Learning

Social phycology, AI and Ethics together can provide valuable models for peer feedback and teamworks

Team Formation

Global economy demands to restructure education to encourage entrepreneurship, creativity and risk-taking. Learning based on teamwork is the path to follow. Within collaborative and task-based education, one of the recurring problems is how to form teams of students. AI allows the analysis of a multitude of factors (sociological, competence, psychological, etc.) to explore the enormous space of possible combinations and find the optimal teams of students in different scenarios and contexts. 

Peer Evaluation

Progress will be made in automatic and peer evaluation processes, which will further democratize education online and throughout life. Advances in natural language processing and computer vision combined with explanation techniques will make the self-assessment that systems provide to students much more informative and useful. Likewise, peer assessment combined with AI techniques will allow the assessment of large groups of online education to be acceptable to teachers. 

Lesson Plans

There are a number of available tools that support teachers in the management of lesson plans on the web. However, none of them is task-centred and support any form of lesson plan's execution over the web. At IIIA, we are interested in the design and execution of these pedagogical workflows. Our Lesson Plans allows to coordinate interactions, ensuring the rules set by the lesson plan are followed, where lesson plans are designed with respect to a selected rubric. Once the lesson plan is defined, a specific graphical user interface (GUI) is automatically generated to allow students navigate through the lesson. Every time the tutor modifies a workflow, a new GUI is generated accordingly without any programming effort. 

Personalised Learning

Hybrid recommender systems and learning analytics allows creating custom-made contents and learning itineraries.

Based on data analytics, Artificial Intelligence algorithms can provide a learning context for the particular needs of students or group of students. We study and create models and algorithms that automatically recommend custom contents and create learning itineraries for the learning needs of students. 

Serious Games

Combine Virtual Reality, AI and gamification to promote learning by playing.

Artificial Intelligence and Virtual Reality provide a rich environment for game-based learning, also called serious games. We develop new personalisation techniques that can be integrated in virtual games to create learning environments where to study and practice several subjects in an inmersive and entretained way.

IIIA develops AI-based software components to offer schools and teachers tools to implement at classrooms. Our toolbox currently offers tools for peer assessments, team composition and lesson plans creation and execution. In what follows, you can play with and test the different demonstrators that shows some of the functionalities offered by our AI-based components.

Team Formation

Cultivation of teamwork, community building, and leadership skills are valuable classroom goals that are more and more introduced at schools. Our aim is to contribute with software technologies that provide teachers with tools to create teams that perform well at diferent levels. 

Synergetic Teams Tool

Partitioning groups of students into competence and cogenial teams for a problem-solving. Eduteams is a Webapp that support the composition of Synergetic teams of students at the classroom.

Congenial Teams Tool 

Partitioning groups of students into gender and psychologically balanced problem-solving teams. Eduteams is a Webapp that support the composition of congenial teams of students at the classroom.

Educational Teams to Companies

Desicion support component to help assign group of students to a Intership project or task. Edu2Com is an Artificial Intelligence component for allocating teams to tasks or projects based on competencies and preferences.

Peer Evaluation

Involving students into accessing others supports teachers but also increase students skills and knowledge. Our aim is to offers computational tools that support the peer assessment in and out of classrooms.

Collaborative Assessment [demo]

Combines teacher and peer assessments to reduce the number of evaluations to make.

Lesson Plans

Our aim is to allow teachers and students to participate into a more flexible, open and collaborative online learning environment. We build tools to support flexible ways to build, share and use collaborative Lesson Plans.

Lesson Plan Editor [demo]

Lesson plan editor to create peer to peer lessons.

Lesson Plan Online Execution [demo]

An online learning environment where executing peer to peer lesson plans.

Filippo Bistaffa
Contract Researcher
Phone Ext. 209

Christian Blum
Scientific Researcher
Phone Ext. 214

Athina Georgara
Industrial PhD Student
Phone Ext. 234

Lissette Lemus del Cueto
Contract Engineer
Phone Ext. 259

Nardine Osman
Tenured Scientist
Phone Ext. 245

Juan A. Rodríguez-Aguilar
Research Professor
Phone Ext. 218

Jordi Sabater-Mir
Tenured Scientist
Phone Ext. 261

Carles Sierra
Research Professor
Phone Ext. 231

Filippo Bistaffa,  Christian Blum,  Jesús Cerquides,  Alessandro Farinelli,  & Juan A. Rodríguez-Aguilar (2021). A Computational Approach to Quantify the Benefits of Ridesharing for Policy Makers and Travellers. IEEE Transactions on Intelligent Transportation Systems, 22, 119-130. [BibTeX]  [PDF]
Marco Schorlemmer,  & Enric Plaza (2021). A Uniform Model of Computational Conceptual Blending. Cognitive Systems Research, 65, 118--137. [BibTeX]  [PDF]
F. A. {Farinelli} (2021). Efficient Coalition Structure Generation via Approximately Equivalent Induced Subgraph Games. IEEE Transactions on Cybernetics, 1-11. [BibTeX]  [PDF]
Jordi Ganzer,  Natalia Criado,  Maite Lopez-Sanchez,  Simon Parsons,  & Juan A. Rodríguez-Aguilar (2020). A model to support collective reasoning: Formalization, analysis and computational assessment. arXiv preprint arXiv:2007.06850. [BibTeX]  [PDF]
Marc Serramia,  Maite Lopez-Sanchez,  & Juan A. Rodríguez-Aguilar (2020). A Qualitative Approach to Composing Value-Aligned Norm Systems. Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems (pp. 1233--1241). [BibTeX]  [PDF]
Jerónimo Hernández-González,  & Jesús Cerquides (2020). A Robust Solution to Variational Importance Sampling of Minimum Variance. Entropy, 22, 1405. [BibTeX]  [PDF]
Francisco Salas-Molina,  Juan A. Rodríguez-Aguilar,  & David Pla-Santamaria (2020). A stochastic goal programming model to derive stable cash management policies. Journal of Global Optimization, 76, 333--346. [BibTeX]  [PDF]
Manel Rodríguez Soto,  Maite López-Sánchez,  & Juan A. Rodríguez-Aguilar (2020). A Structural Solution to Sequential Moral Dilemmas. Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems (pp. 1152--1160). [BibTeX]  [PDF]
Anna Puig,  Inmaculada Rodríguez,  Josep Ll Arcos,  Juan A. Rodríguez-Aguilar,  Sergi Cebrián,  Anton Bogdanovych,  Núria Morera,  Antoni Palomo,  & Raquel Piqué (2020). Lessons learned from supplementing archaeological museum exhibitions with virtual reality. Virtual Reality, 24, 343--358. [BibTeX]  [PDF]
Nardine Osman,  Carles Sierra,  Ronald Chenu-Abente,  Qiang Shen,  & Fausto Giunchiglia (2020). Open Social Systems. Nick Bassiliades, Georgios Chalkiadakis, & Dave Jonge (Eds.), Multi-Agent Systems and Agreement Technologies (pp. 132--142). Springer International Publishing. [BibTeX]  [PDF]
Filippo Bistaffa,  Juan A. Rodríguez-Aguilar,  & Jesús Cerquides (2020). Predicting Requests in Large-Scale Online P2P Ridesharing. arXiv preprint arXiv:2009.02997. [BibTeX]  [PDF]
Jesús Cerquides,  Juan A. Rodríguez-Aguilar,  Rémi Emonet,  & Gauthier Picard (2020). Solving Highly Cyclic Distributed Optimization Problems Without Busting the Bank: A Decimation-based Approach. Logic Journal of the IGPL. [BibTeX]
Dave de Jonge,  & Dongmo Zhang (2020). Strategic negotiations for extensive-form games. Autonomous Agents and Multi-Agent Systems, 34. [BibTeX]  [PDF]
Athina Georgara,  Carles Sierra,  & Juan A. Rodríguez-Aguilar (2020). TAIP: an anytime algorithm for allocating student teams to internship programs. arXiv preprint arXiv:2005.09331. [BibTeX]  [PDF]
Jesús Vega,  M. Ceballos,  Josep Puyol-Gruart,  Pere García,  B. Cobo,  & F. J. Carrera (2020). TES X-ray pulse identification using CNNs. ADASS XXX . [BibTeX]  [PDF]
Marta Poblet,  & Carles Sierra (2020). Understanding Help as a Commons. International Journal of the Commons, 14, 281--493. [BibTeX]  [PDF]
Antoni Perello-Moragues,  & Pablo Noriega (2020). Using Agent-Based Simulation to Understand the Role of Values in Policy-Making. Harko Verhagen, Melania Borit, Giangiacomo Bravo, & Nanda Wijermans (Eds.), Advances in Social Simulation (pp. 355--369). Springer International Publishing. [BibTeX]
Paula Chocron,  & Marco Schorlemmer (2020). Vocabulary Alignment in Openly Specified Interactions. Journal of Artificial Intelligence Research, 68, 69--107. [BibTeX]
Mariela Morveli Espinoza,  J.C. Nieves,  A. Possebom,  Josep Puyol-Gruart,  & C.A. Tacla (2019). An argumentation-based approach for identifying and dealing with incompatibilities among procedural goals. International Journal of Approximate Reasoning, 105, 1 - 26. [BibTeX]
Mariela Morveli Espinoza,  Ayslan Trevizan Possebom,  Josep Puyol-Gruart,  & C.A Tacla (2019). Argumentation-based intention formation process. DYNA, 86, 82 - 91. [BibTeX]
Francisco Salas-Molina,  Juan A. Rodríguez-Aguilar,  & David Pla-Santamaria (2019). Characterizing compromise solutions for investors with uncertain risk preferences. Operational Research, 19, 661--677. [BibTeX]  [PDF]
Marc Serramia,  Jordi Ganzer-Ripoll,  Maite López-Sánchez,  Juan A. Rodríguez-Aguilar,  Natalia Criado,  Simon Parsons,  Patricio Escobar,  & Marc Fernández (2019). Citizen Support Aggregation Methods for Participatory Platforms.. CCIA (pp. 9--18). [BibTeX]
Jordi Ganzer-Ripoll,  Natalia Criado,  Maite Lopez-Sanchez,  Simon Parsons,  & Juan A. Rodríguez-Aguilar (2019). Combining social choice theory and argumentation: Enabling collective decision making. Group Decision and Negotiation, 28, 127--173. [BibTeX]  [PDF]
Karla Trejo,  Pere García,  & Josep Puyol-Gruart (2019). Metadata Generation for Multi-Text Classification in Structured Data. Artificial Intelligence Research and Development (pp. 417-421). [BibTeX]
Antoni Perello-Moragues,  Pablo Noriega,  & Manel Poch (2019). Modelling Contingent Technology Adoption in Farming Irrigation Communities. Journal of Artificial Societies and Social Simulation, 22, 1. [BibTeX]
Francisco Salas-Molina,  Juan A. Rodríguez-Aguilar,  David Pla-Santamaria,  & Ana García-Bernabeu (2019). On the formal foundations of cash management systems. Operational Research, 1--15. [BibTeX]  [PDF]
Inmaculada Rodriguez,  Anna Puig,  Juan A. Rodríguez-Aguilar,  Josep Lluis Arcos,  Sergi Cebrián,  Anton Bogdanovych,  Núria Morera,  Raquel Piqué,  & Antoni Palomo (2019). On the Relationship between Subjective and Objective Measures of Virtual Reality Experiences: a Case Study of a Serious Game. International Symposium on Gamification and Games for Learning (GamiLearn 2019) . [BibTeX]
Francisco Salas-Molina,  Juan A. Rodríguez-Aguilar,  & David Pla-Santamaria (2019). On the use of multiple criteria distance indexes to find robust cash management policies. INFOR: Information Systems and Operational Research, 57, 345-360. [BibTeX]
Marc Serramia,  Maite López-Sánchez,  Juan A. Rodríguez-Aguilar,  & Patricio Escobar (2019). Optimising Participatory Budget Allocation: The Decidim Use Case. Artificial Intelligence Research and Development (pp 193-202). IOS Press. [BibTeX]
Juan Carlos Teze,  Antoni Perelló-Moragues,  Lluís Godo,  & Pablo Noriega (2019). Practical reasoning using values: an argumentative approach based on a hierarchy of values. Annals of Mathematics and Artificial Intelligence, 293-319. [BibTeX]  [PDF]
Ewa Andrejczuk,  Filippo Bistaffa,  Christian Blum,  Juan A. Rodríguez-Aguilar,  & Carles Sierra (2019). Synergistic team composition: A computational approach to foster diversity in teams. Knowledge-Based Systems, 182. [BibTeX]
Dave Jonge,  Tim Baarslag,  Reyhan Aydoğan,  Catholijn Jonker,  Katsuhide Fujita,  & Takayuki Ito (2019). The Challenge of Negotiation in the Game of Diplomacy. Marin Lujak (Eds.), Agreement Technologies 2018, Revised Selected Papers (pp. 100-114). Springer International Publishing. [BibTeX]
Carles Sierra,  Nardine Osman,  Pablo Noriega,  Jordi Sabater-Mir,  & Antoni Perello-Moragues (2019). Value alignment: A formal approach. Responsible Artificial Intelligence Agents Workshop (RAIA) in AAMAS 2019 . [BibTeX]  [PDF]
Manfred Eppe,  Ewen Maclean,  Roberto Confalonieri,  Oliver Kutz,  Marco Schorlemmer,  Enric Plaza,  & Kai-Uwe Kühnberger (2018). A computational framework for conceptual blending. Artificial Intelligence, 256, 105-129. [BibTeX]
Filippo Bistaffa,  & Alessandro Farinelli (2018). A COP Model For Graph-Constrained Coalition Formation. Journal of Artificial Intelligence Research, 62, 133-153. [BibTeX]
Filippo Bistaffa,  & Alessandro Farinelli (2018). A COP Model for Graph-Constrained Coalition Formation (Extended Abstract). International Joint Conference on Artificial Intelligence (IJCAI-ECAI 2018) (pp. 5553-5557). AAAI Press. [BibTeX]
Pham Tran Anh Quang,  Kamal Deep Singh,  Juan A. Rodríguez-Aguilar,  Gauthier Picard,  Kandaraj Piamrat,  Jesús Cerquides,  & César Viho (2018). AD3-GLaM: A Cooperative Distributed QoE-based Approach for SVC Video Streaming over Wireless Mesh Networks. Ad Hoc Networks, 80, 1-15. [BibTeX]
Fèlix Bou,  Enric Plaza,  & Marco Schorlemmer (2018). Amalgams, colimits, and conceptual blending. Concept Invention: Foundations, Implementation, Social Aspects and Applications. Springer. [BibTeX]
Francisco Salas-Molina,  David Pla-Santamaria,  & Juan A. Rodríguez-Aguilar (2018). A multi-objective approach to the cash management problem. Annals of Operations Research, 267, 515-529. [BibTeX]
Filippo Bistaffa,  Juan A. Rodríguez-Aguilar,  Jesús Cerquides,  & Christian Blum (2018). A Simulation Tool for Large-Scale Online Ridesharing (Demonstration). International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2018) (pp. 1797-1799). IFAAMAS. [BibTeX]
  • UNESCO Declaration. In May 2019, around 100 UNESCO member states made a series of recommendations that mark the way forward in the coming years. The first and most relevant is that AI has to be integrated into the education system. AI must be taught and at the same time used to strengthen student learning. This integration and use must be based on scrupulous respect for human rights. It must serve to train students with a critical spirit regarding the use of this technology that allows them to understand the risks and take advantage of the opportunities it offers us. The future of AI in the educational world is fascinating.  
  • SQUIRREL AI. An online education company specialising in intelligent adaptive education.