About me

I am a postdoctoral fellow working with Milind Tambe at Harvard to build AI systems with positive social impact!

I completed my Ph.D. at the Artificial Intelligence Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), under the supervision of Professor Boi Faltings (my thesis is nominated for distinction).

Prior to that, I studied Electrical and Computer Engineering at the National Technical University of Athens (NTUA) after which I was employed as a research associate at the Microprocessors and Digital Systems Laboratory (microLAB).

My Research

My research interests lie in the areas of multi-agent systems (cooperation & coordination), (deep reinforcement) learning, and game theory. I have always been strongly motivated by the goal to provide practical solutions for real problems. My work aims to facilitate the integration of intelligent agents in the society by building systems that work in the real-world.

The true potential of intelligent systems will only be realized when they are tightly integrated into our daily lives. Such a future will involve a plethora of interacting autonomous learning agents mediating many aspects of our social, economic, cultural, and technological world, giving rise to complex machine behavior. Carefully designing autonomous learning systems and studying their behavior is, I believe, one of the most pressing problems we currently face in AI.

Publications

Working Papers

• "A Practical Influence Approximation for Privacy-Preserving Data Filtering in Federated Learning",
  Lj. Rokvic, P. Danassis, B. Faltings
  [ Paper | Arxiv ]

Journal Papers

2022

• "Putting Ridesharing to the Test: Efficient and Scalable Solutions and the Power of Dynamic Vehicle Relocation",
  P. Danassis, M. Sakota, A. Filos-Ratsikas, B. Faltings,
  Artificial Intelligence Review, Springer [ Paper | Arxiv]

• "Exploiting Environmental Signals to Enable Policy Correlation in Large-scale Decentralized Systems",
  P. Danassis, Z. D. Erden, B. Faltings,
  Journal of Autonomous Agents and Multi-agent Systems, Springer [ Paper ]

Conference Papers

2023

• "AI-driven Prices for Externalities and Sustainability in Production Markets",
  P. Danassis, A. Filos-Ratsikas, H. Chen, M. Tambe, B. Faltings
  AAMAS 2023: Proceedings of the 22nd International Conference on Autonomous Agents and MultiAgent Systems [ Paper | Arxiv | Code ]

2022

• "A Distributed Differentially Private Algorithm for Resource Allocation in Unboundedly Large Settings",
  P. Danassis, A. Triastcyn, B. Faltings,
  AAMAS 2022: Proceedings of the 21st International Conference on Autonomous Agents and MultiAgent Systems [ Paper | Arxiv ]

2021

• "Improving Multi-agent Coordination by Learning to Estimate Contention",
  P. Danassis, F. Wiedemair, B. Faltings,
  IJCAI 2021: Proceedings of the 30th International Joint Conference on Artificial Intelligence [ Paper | Arxiv ]

• "Improved Cooperation by Exploiting a Common Signal",
  P. Danassis, Z. D. Erden, B. Faltings,
  AAMAS '21: Proceedings of the 20th International Conference on Autonomous Agents and MultiAgent Systems [ Paper | Video | Arxiv | Code ]

2020

• "Efficient Allocations in Constant Time: Towards Scalable Solutions in the Era of Large Scale Intelligent Systems",
  P. Danassis, B. Faltings,
  ECAI 2020: Proceedings of the 24th European Conference on Artificial Intelligence, 2-page Highlight Paper [ Paper | Poster ]

2019

• "Anytime Heuristic for Weighted Matching Through Altruism-Inspired Behavior",
  P. Danassis, A. Filos-Ratsikas, B. Faltings,
  IJCAI 2019: Proceedings of the 28th International Joint Conference on Artificial Intelligence [ Paper | Arxiv | Poster ]

• "Courtesy as a Means to Coordinate",
  P. Danassis, B. Faltings,
  AAMAS '19: Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems [ Paper | Arxiv | Poster ]

Publications Prior to my Doctoral Studies

2017

• "Supporting Decision Making for Large-Scale IoTs: Trading Accuracy with Computational Complexity",
  K. Siozios, P. Danassis, N. Zompakis, C. Korkas, E. Kosmatopoulos and D. Soudris,
  Components and Services for IoT Platforms: Paving the Way for IoT Standards, Springer (invited book chapter) [ Pdf | Doi ]

• "A Low-Complexity Control Mechanism Targeting Smart Thermostats",
  P. Danassis, K. Siozios, C. Korkas, D. Soudris, E. Kosmatopoulos,
  Energy and Buildings, Elsevier [ Pdf | Doi ]

2016

• "ANT3D: Simultaneous Partitioning and Placement for 3-D FPGAs based on Ant Colony Optimization",
  P. Danassis, K. Siozios, D. Soudris,
  IEEE Embedded Systems Letters, IEEE [ Pdf | Doi ]

• "Parallel Application Placement onto 3-D Reconfigurable Architectures",
  P. Danassis, K. Siozios, D. Soudris,
  International Conference on Modern Circuits and Systems Technologies (MOCAST), IEEE sponsored [ Pdf | Doi ]

Ph.D. Thesis

• 2022, P. Danassis, "Scalable Multi-agent Coordination and Resource Sharing" [ Pdf ]

Diploma Thesis

• 2015, P. Danassis, "A Novel 3-D FPGA Placement Algorithm based on Ant Colony Optimization" [ English | Greek ]

Peer-Reviewed Workshops and Symposia

• 2022, "Privacy-preserving Data Filtering in Federated Learning Using Influence Approximation", Lj. Rokvic, P. Danassis, B. Faltings, FL-NeurIPS'22: International Workshop on Federated Learning: Recent Advances and New Challenges at NeurIPS 2022 [ Paper ]
• 2022, "Privacy-preserving Data Filtering in Federated Learning Using Influence Approximation", Lj. Rokvic, P. Danassis, B. Faltings, FL-IJCAI'22: International Workshop on Trustworthy Federated Learning at IJCAI 2022 [ Paper ]
• 2021, "Differential Privacy Meets Maximum-weight Matching", P. Danassis, A. Triastcyn, B. Faltings, ALA 2021: Adaptive Learning Agents Workshop at AAMAS [ Paper ]
• 2021, "Differential Privacy Meets Maximum-weight Matching", P. Danassis, A. Triastcyn, B. Faltings, AASG 2021: Autonomous Agents for Social Good Workshop at AAMAS [ Paper ]
• 2021, "Improved Cooperation by Exploiting a Common Signal", P. Danassis, Z. D. Erden, B. Faltings, AASG 2021: Autonomous Agents for Social Good Workshop at AAMAS [ Paper ]
• 2021, "Differential Privacy Meets Maximum-weight Matching", P. Danassis, A. Triastcyn, B. Faltings, PPAI 2021: The Second AAAI Workshop on Privacy-Preserving Artificial Intelligence [ Paper ]
• 2020, "Learning to Persist or Switch: Efficient and Fair Allocations in Large-scale Multi-agent Systems", P. Danassis, B. Faltings, ALA 2020: Adaptive Learning Agents Workshop at AAMAS [ Paper ]
• 2019, "Anytime Heuristic for Weighted Matching Through Altruism-Inspired Behavior", P. Danassis, A. Filos-Ratsikas, B. Faltings, ALA 2019: Adaptive Learning Agents Workshop at AAMAS [ Paper ]
• 2018, "Courtesy as a Means to Anti-coordinate", P. Danassis, B. Faltings, ALA 2018: Adaptive Learning Agents Workshop at AAMAS [ Paper ]
• 2018, "Learning in Ad-hoc Anti-coordination Scenarios", P. Danassis, B. Faltings, AAAI Spring Symposium Series [ Paper ]