The 39th Annual AAAI Conference on Artificial Intelligence
February 25 – March 4, 2025 | Philadelphia, Pennsylvania, USA
AAAI-25 Bridge Program
Sponsored by the Association for the Advancement of Artificial Intelligence
February 25-26, 2025 | Pennsylvania Convention Center | Philadelphia, Pennsylvania, USA
B1: AI for Medicine and Healthcare
B2: Bridge between AI and Scientific Knowledge Organization
B3: Bridging Cognitive Science and AI to Bridge Neuro and Symbolic AI
B4: Bridging Planning and Reasoning in Natural Languages with Foundational Models (PLAN-FM)
B1: AI for Medicine and Healthcare
Our AAAI bridge program aims to address the critical gap between the impressive capabilities of AI in medical research and its limited integration into real-world clinical practice. While AI technologies, such as computer vision algorithms for medical imaging and natural language processing for patient data analysis, have demonstrated great potential, human supervision is often required to ensure safety and reliability.
A key challenge is the disconnect between clinicians, who may lack the technical expertise to fully apply AI tools, and AI developers, who may not completely understand clinical workflows and needs. Our bridge program seeks to bring these two groups together.
Through this initiative, clinicians will gain better insights into how AI can enhance clinical outcomes, while AI developers will learn to design solutions that align with the practical demands of clinical environments. This collaborative effort will foster stronger connections and pave the way for the integration of AI into medical practice.
For further and latest information please check our AIMedHealth website: https://sites.google.com/view/aimedhealth-aaai/home
Format
This event will be one day long. It will include presentation session, training session, workshop session, poster session and a panel talk. We will also invite industry partners to share their perspectives on AI in the medical field. Full schedule can be found: https://sites.google.com/view/aimedhealth-aaai/schedule
Topics
This bridge event aims to integrate techniques from the AI field with insights from clinical practice. By combining expertise from both domains, the bridge will address critical challenges in effectively and safely integrating AI into medical settings. We invite submissions that explore the technical aspects or the clinical perspective of this integration. The following topics are examples, but we encourage a wide range of related contributions to AI for medicine and AI for healthcare. More topic details can be found: https://sites.google.com/view/aimedhealth-aaai/calls.
Submissions
We encourage two formats of submissions:
- Extended Abstracts: 2-4 pages (exclude reference)
- Full-Length Papers: Up to 8 pages (exclude reference)
AAAI style template: https://aaai.org/conference/aaai/aaai-25/
Link to Submissions: https://easychair.org/conferences/?conf=aimedhealth25
Important Dates
Submission Deadline: November 25, 2024
Notification of Acceptance: December 9, 2024
Early Registration Deadline: December 19, 2024
Bridge Committee
- Junde Wu, University of Oxford, junde.wu@linacre.ox.ac.uk
- Jiayuan Zhu, University of Oxford, jiayuan.zhu@keble.ox.ac.uk
- Min Xu, Carnegie Mellon University, mxu1@cs.cmu.edu
- Yueming Jin, National University of Singapore, ymjin@nus.edu.sg
- Alex Novak, Oxford University Hospitals NHSFT, alex.novak@ouh.nhs.uk
- Sarim Ather, Oxford University Hospitals NHSFT, Sarim.Ather@ouh.nhs.uk
- Bartek Papiez, University of Oxford, bartlomiej.papiez@bdi.ox.ac.uk
- Alison Noble, University of Oxford, alison.noble@eng.ox.ac.uk
B2: First AAAI Bridge on Artificial Intelligence for Scholarly Communication
Description
The scientific community of today faces the problem of scientific papers overload in their respective domains. There is an increasingly large number of currently 3 million papers published every year in addition to the approximatively 200 million ones already published. This gives rise to the research question: “How can we provide a reliable and living scientific knowledge base that empowers researchers to query, synthesize and analyse the vast body of scholarly knowledge?” Although many tools for assisting researchers during scientific knowledge extraction and organization exist, it has been reported that most researchers continue to depend on manual methods. Artificial intelligence (AI) for scholarly communication (AI4SC) aims to leverage AI methodologies, models and applications for scientific knowledge extraction, organisation and use.
The AI4SC bridge program aims to bring together a broad audience (from different disciplines) of students, researchers and AI-experts actively developing/using or not AI-artifacts such as datasets, connectionist and symbolic AI models for scientific knowledge extraction, organization and use to identify common problems, facilitate collaborations and define future research directions. To this end, the following research questions will be considered:
- How to acquire scientific knowledge from research papers? The aim of this question is to document methodologies, methods, and tools being used for scientific knowledge acquisition. Participants will be invited to describe machine learning (ML) models used during the scientific knowledge acquisition process and datasets used to train these models.
- How to organise scientific knowledge? Once extracted, scientific knowledge should be organized in such a way that fellow researchers can benefit from it. Participants will describe AI-based models (connectionist and symbolic models) and tools used for scientific knowledge organization.
- How to improve the usability of tools for knowledge extraction and organization? Tools for knowledge extraction and organization are used in diverse research disciplines including social sciences, engineering and technology, education, environmental sciences, and business and management. Participants will be invited to describe how user interfaces are used for making easy access to scientific knowledge. Thereafter, a panel discussion will be held to discuss how to make these tools more accessible to researchers in all domains.
Given that LLMs is the state-of-the-art in several NLP tasks, a particular attention will be oriented towards the use of LLMs for scientific knowledge extraction and organization.
For more information, visit the following web site: https://sites.google.com/view/ai4sc/edition/ai4sc-AAAI2025 or email us with questions at ai4schocom@gmail.com.
Format
The bridge will include introductory lectures, tutorials, hand on session, panel discussion, demos and poster sessions. The poster session will be devoted to AI-artifacts for scholarly communication.
Submissions
Participants are invited to submit papers of two pages max, excluding references and appendices using Open Review: https://openreview.net/group?id=AAAI.org/2025/Bridge/AI4SC. Given that LLMs is the state-of-the-art in several NLP tasks, a particular attention will be oriented towards the use of LLMs for scientific knowledge extraction, organization or exploitation. The papers should be formatted in the AAAI two column style (see https://aaai.org/authorkit24-2) and be anonymised. Papers that have been submitted/published in other conferences/journals are also welcome.
Participants are invited to submit in the following tracks:
- Poster track is dedicated to students and researchers who want to present how they exploit existing literature in their work. A particular attention will be oriented towards the use of AI in scientific communication.
- Tutorial Track is dedicated to provide participants with practical skills, tools or detailed knowledge in the use of AI in scientific communication. Experts or experienced professionals in the field are invited to submit tutorials including live demonstrations, interactive exercises, or guided practice.
- Dataset Track aims to showcase new, high-quality datasets and discuss their structure, methodology, and use. The participants are invited to submit papers describing datasets that have potential value for research, analysis, or application of AI for scholarly communication. These datasets can be new or updated ones. Participants should provide the sources of data, describe how the data was collected, cleaned, structured and potential applications.
- System Track aims to showcase how systems, frameworks, tools or platforms that have been developed for AI in scientific communication work, their design, and their real-world applications. The participants should submit papers describing fully implemented and functional systems.
- NB: participants can submit to more than one track.
Important Dates
Submissions due: December 10th 2024
Notification to authors: December 20th 2024
Bridge @ AAAI: February, 25-26
Organizers
- Dr. Azanzi Jiomekong, University of Yaounde
- Prof. Dr. Sören Auer, Leibniz Universität Hannover
B3: Bridging Cognitive Science and AI to Bridge Neuro and Symbolic AI
Description
This bridge aims to foster a structural and long-lasting connection between cognitive science and AI, focusing on how cognitive mechanisms can inspire neuro-symbolic AI. The objective is to explore how cognitive science, with its insights into human reasoning (both intuitive and deliberate), can inform the integration of data-driven (neuro) and symbolic AI approaches. The ultimate goal is to develop AI systems capable of higher-level reasoning, introspection, and trustworthiness—characteristics that current AI systems lack.
Topics
We invite submissions that address topics related to:
- Cognitive science contributions to AI architecture and design
- Integration of neural and symbolic AI
- Knowledge representation and reasoning in AI
- Large Language Models (LLMs) and their cognitive inspirations
- Meta-cognition in AI: how AI systems can self-reflect on their processes and decisions
Format of Bridge
The bridge will be organized into a one-day event, featuring a combination of:
- Morning Tutorials/Labs: Three interactive sessions to introduce key cognitive science concepts relevant to AI (3 sessions, 1 hour each)
- Invited Talks: Two talks from leading experts in cognitive science and AI (1 hour each)
- Paper Presentations: Selected submissions will be presented in two sessions (1 hour each)
- Panel Discussion: A concluding panel discussion (1.5 hours) featuring key figures from cognitive science and AI to discuss the future of neuro-symbolic AI
Attendance
All presentations must be in person. Only virtual attendance will be permitted and available to everyone registered for the event. There will not be any session recordings shared after the event.
Submission Requirements
We invite submitting original research papers up to 8 pages plus additional pages solely for references. Papers should present innovative solutions, theoretical perspectives, or open challenges relevant to bridging cognitive science and AI.
Papers must be formatted in AAAI two-column, camera-ready style; see the AAAI-25 author kit for details.
Submissions are anonymous and must conform to the instructions (detailed below) for double-blind review.
Bridge Submissions deadline: Sunday, November 24, 2024
Notifications Sent to Authors: Monday, December 9, 2024
Submission Site Information
For questions or issues with submission, please email: andrea.loreggia[AT]unibs.it
Bridge Chair
Chair: Francesca Rossi <Francesca.Rossi2@ibm.com>,
Kenneth D Forbus <forbus@northwestern.edu>,
Andrea Loreggia <andrea.loreggia@gmail.com>,
Nicholas Mattei <nsmattei@gmail.com>,
Oscar Romero<oscar.j.romero@gmail.com>,
Brent Venable<brent.venable@gmail.com>,
Biplav Srivastava <BIPLAV.S@sc.edu>
Bridge URL
More detailed information, including the program schedule and speakers, will be available at: https://sites.google.com/view/cosainsai
B4: Bridging Planning and Reasoning in Natural Language with Foundational Models (PLAN-FM)
Description
There is a growing interest in utilizing Foundational Models for complex tasks that require multi-step reasoning and planning. This promising area of research is seeing an increasing number of contributions from researchers in the fields of Natural Language Processing (NLP), Planning, and Robotics. PLAN-FM Bridge program facilitates collaboration and knowledge-sharing among these researchers. This program will provide a stage to discuss and exchange perspectives, identify critical challenges and outline research agendas.
Objective
In this bridge, we aim to foster interactions between NLP, Planning and Robotics researchers that are using and working with foundational models, provide a stage to discuss and exchange common terminology, identify a shared research agenda and pinpoint the most critical challenges, and create a rich repository of resources that can be leveraged by the community, thereby facilitating cross-pollination of ideas and fostering collaboration across research fields to drive the robust advancement of automated planning.
Topics/Open Questions
- How can we effectively harness the power of foundational models for planning purposes?
- What are some critical decision-making challenges that can be addressed by leveraging foundational models?
- What kind of guarantees can we expect when using foundational models for planning?
- How can we overcome their limited reliability before deploying them in real-world applications?
- Can we develop a comprehensive suite of datasets and benchmarks that can be shared across communities to evaluate planning abilities in a consistent and reliable manner?
- What are the key evaluation considerations and metrics that should be used to assess the reliability of planning approaches, and what tools can be leveraged across communities to facilitate this evaluation?
- What are the characteristics in future foundation models that can help planning?
Format of Bridge
The bridge will provide a rich program of activities, including tutorials, panel discussions, talks, and networking opportunities. The bridge event will include a half-day tutorial session with a goal of fostering interdisciplinary understanding and exchange of fundamental principles. Second half of the day will include invited talks, panel, and poster session.
We also accept submissions of abstracts, demonstrations as well as position papers.
Submission Requirement
We solicit submissions relevant to the bridge program of the following types:
- System Demonstration – up to 4 pages (include description of the demo and a screenshot or link)
- Position papers – up to 4-8 pages (excluding references)
- Abstracts – up to 2 pages (excluding references)
Papers must be formatted in AAAI two-column, camera-ready style; see the AAAI-25 author kit for details.
Important Dates:
- Paper submission deadline: Sunday, November 24, 2024 (AOE)
- Paper acceptance notification: Monday, December 9, 2024 (AOE)
Refer to the bridge website for latest information: https://plan-fm.github.io/
Submission Site Information:
Paper submissions should be made through easychair: https://easychair.org/conferences/?conf=planfm2025
Bridge Chairs
- Harsha Kokel, IBM Research
- Shirin Sohrabi, IBM Research
- Soham Dan, Microsoft
- Manling Li, Northwestern University
- Yu Su, Ohio State University
Advising Committee:
- Biplav Srivastava, University of South Carolina
- Sriraam Natarajan, University of Texas at Dallas
- Subbarao Khambampati, Arizona State University
Please send your inquiries to plan-fm-bridge@googlegroups.com
Bridge URL:
B5: Collaborative AI and Modeling of Humans
Advances in Artificial Intelligence (AI) methods have yield unprecedented results, even surpassing human performance on a variety of well-defined tasks. However, most real-world problems, especially those involving humans, are complex, multi-dimensional and hard to specify a priori. A principled way to address this limitation is to allow AI systems to collaborate with humans, and thereby actively anticipate and adapt to humans’ needs and abilities. To enable such reasoning, AI must be equipped with computational models of human behavior. Such models have been heavily investigated in cognitive science and AI-adjacent fields such as human-AI interaction, human-computer interaction (HCI), and behavioral game theory. However, due to differences in research goals and experimental settings, these communities have operated more or less independently, with limited exchange of theories and methods.
Following a successful first edition in 2024, we aim to bring together members of the communities relevant to human-AI collaboration and user modeling to exchange theories, perspectives, and methods.
Topics
The space of disciplines covered by the relevant fields is very large and submissions are expected to cover topics such as:
- Machine learning with human(s) in the loop
- User modeling, theory of mind, and computational rationality
- Human-AI collaboration
Format
This event will be one day long. It will start with two keynote talks, from the perspectives on either side of the bridge topic of human modeling in AI. Next, a tutorial will provide a deep dive into the bridge topic. This will be followed by a poster session where authors of accepted papers will be invited to present their work. The day will conclude with an interactive discussion with a panel of experts with ample time to discuss.
Submissions
We encourage submissions for the poster session on all topics relevant to the bridge but expect they include a dedicated section elucidating the potential interconnection of both disciplines. Submissions are reviewed double-blind, so they should be anonymized. There will be no proceedings, so papers that have been or will be submitted or published in other conferences or journals are also welcome.
We accept papers of 2 to 8 pages, excluding references and appendices. The papers should be formatted in the AAAI two-column, camera-ready style (see https://aaai.org/authorkit25 for details) and authors can submit their works through https://openreview.net/group?id=AAAI.org/2025/Bridge/CAIHU.
For more detailed submission instructions, please visit https://sites.google.com/view/caihu25/call-for-papers?authuser=0.
Target Audience
In this bridge program, we hope to bring together a broad audience of students, researchers, and practitioners in fields relevant to human-AI collaboration and human behavior modeling including AI, HCI, and CogSci.
Organizers
- Andrew Howes; University of Exeter, England; A.Howes2@exeter.ac.uk
- Samuel Kaski; Aalto University, Finland and University of Manchester, UK; samuel.kaski@aalto.fi
- Frans A. Oliehoek; Delft University of Technology, Netherlands; f.a.oliehoek@tudelft
- Nuria Oliver; ELLIS Alicante, Spain; nuria@ellisalicante.org
- Matthew E. Taylor; University of Alberta & Alberta Machine Intelligence Institute, Canada; matthew.e.taylor@ualberta.ca
Additional information
For more information visit https://sites.google.com/view/caihu25/home?authuser=0 or email us with questions at caihu.aaai@gmail.com.
B6: Combining AI and ORMS for Better Trustworthy Decision Making
Description
Artificial Intelligence (AI), including Generative AI, and Operations Research/Management Science (OR/MS) offer proven but distinct approaches to decision-making with data and models. However, challenges persist in applying them to vital socio-technical environments where human and artificial systems interact. These include the need to combine AI and OR/MS for the best solution, aligning models with human values and promoting trust, and the expertise and time required for their application, which limit wider use.
Main Objectives
The goal of this bridge program is to unite AI and OR/MS practitioners and researchers to improve trustworthy decision-making in key socio-technical areas such as supply chains, healthcare, crisis management, homeland security, robotics, wildlife conservation, medicine, transportation, and finance. It aims to equip them with better tools by familiarizing them with each other’s techniques and domains and bring the disciplines together to advance the research and applications at the intersection of AI and OR/MS so as to improve decision-making.
Topics
- Utilizing AI, OR/MS, and their integration for decision-making.
- Exploring current state-of-the-art research and identifying new directions in combining AI and OR/MS for improved trustworthy decision-making, including integrating Large Language Models with OR/MS and other AI tools to democratize advanced decision-making capabilities and integrating OR/MS and AI to improve trustworthiness.
- Identifying key domains and use cases where AI and OR/MS can improve decision-making.
Format of the Bridge
This two-day bridge will include: a half day of invited talks and a tutorial on using OR/MS for decision-making; a half day of submitted presentations and posters on applying AI and/or OR/MS and state-of-the-art research on their integration for trustworthy decision-making; a half day of presentations on future research directions and key domains for AI and OR/MS integration; and a half day of discussions to refine research priorities and focus use cases, and define next steps.
Attendance
Attendance is open to all interested students, practitioners, and researchers.
Submission Requirements
Participants interested in presenting can submit a presentation proposal about one or more of the following:
- An application of AI and/or OR/MS (individually or together) to decision-making.
- Existing research integrating AI and OR/MS for decision-making.
- New, important research directions that integrate AI and OR/MS for decision-making.
- Decision-making domains and use cases where AI and OR/MS should be jointly applied, with a rationale for the combined approach.
- Relevant surveys presentation.
Submissions can range from a one-page abstract to a full journal article and may include work at any level of maturity, as well as previously published work.
Submission Site Information: https://easychair.org/conferences/?conf=aiorms2025
Bridge Committee:
- Sven Koenig, University of California, Irvine, sven.koenig@uci.edu
- Michela Milano, Università di Bologna, michela.milano@unibo.it.
- Willem-Jan van Hoeve, Carnegie Mellon University, vanhoeve@andrew.cmu.edu.
- Segev Wasserkrug, IBM Research and Technion, segevw@il.ibm.com.
Bridge External URL: https://aaai.org/conference/aaai/aaai-25/bridge-ai-orms/
B7: Constraint Programming and Machine Learning
Bringing together Constraint Programming (CP) and Machine Learning (ML) is an important aspect of the larger goal of integrating Reasoning and Learning. Participants are not expected to have prior experience in both fields, but to have familiarity with each at least at the level of an introductory AI course. The Bridge is designed to educate and to build community, to provide opportunities to interact, discuss, raise awareness and find collaborators.
Focus
The focus of this one-day Bridge will be on bringing together the traditional AI fields of constraint-based reasoning and machine learning, but participants from related fields of reasoning, optimization and learning, e.g. SAT, operations research, data mining, will be welcome.
Submissions
You can submit in any of a variety of Tracks. There are many Tracks. We do not necessarily expect to receive submissions for every Track, but we wish to maximize opportunities and options for contributing to the Bridge and the Bridge community. You may submit to more than one Track. The simplest option is the Introductions Track, which has minimal requirements, and provides an opportunity for participants to introduce themselves, with a view to facilitating interaction and enabling collaboration, during the Bridge day and afterwards. Note that if we receive too many submissions to this Track to accommodate for physical presentation in the time available, appropriate participants will be chosen for presentation on a first come first served basis, so you are encouraged to submit early.
The full list of Track options is available at the CPML Bridge website, along with full submission requirements and instructions and other important information. Submissions will be through EasyChair.
Important Dates
- November 25, 2024: Bridge Submissions Due
- December 9, 2024: Notifications Sent to Authors
- December 19, 2024: AAAI Early Registration Deadline
Organizers
Eugene Freuder. UCC. eugene.freuder@insight-centre.org
Barry O’Sullivan. UCC. barry.osullivan@insight-centre.org
Steering Committee
Christian Bessiere (U.Montpellier, LIRMM)
Luc De Raedt (KU Leuven)
Eugene Freuder (University College Cork)
Tias Guns (KU Leuven)
Kevin Leyton-Brown (Uni. of British Columbia)
Michela Milano (University of Bologna)
Nina Narodytska (VMware Research, USA)
Barry O’Sullivan (University College Cork)
B8: Continual Causality
The fields of causality and continual learning investigate complementary aspects of human cognition, and artificial intelligence must emulate both if it is to reason and generalize in complex environments. On the one hand, causality theory provides the language, algorithms, and tools to discover and infer cause-and-effect relationships from data. On the other hand, continual learning systems balance learning from new data as they become available with retaining previous knowledge while experiencing distribution shift over time. Our recurring “Continual Causality” bridge continues working towards a unified treatment of these fields by providing a space to learn and discuss, and to connect and build a diverse long-term community.
Topics
We invite submissions that present general positions or visions of how to link the two fields, outline challenges that need to be overcome, highlight synergies, or discuss first practical approaches and solutions to relevant problems. Our vision is for the community to voice diverse views that have the potential to advance AI through an ongoing cross-disciplinary exchange.
Format of Bridge
Our two day bridge is composed of a wide range of activities, including traditional tutorials on the educational side, invited vision talks and contributed ones based on submitted papers, interactive sessions in the form of a speakers panel, community breakout discussions, and a challenge session.
Attendance
We invite contributions in the form of original papers, recently published works on the bridge’s theme, and challenge submissions. All attendees of AAAI-25 are further invited to actively participate in our discussion and breakout sessions.
Submission Requirements
Submission can either be non-archival or for inclusion in a Proceedings of Machine Learning Research (PMLR) volume. However, all works must be original and limited to four pages (excluding references and optional appendices) in the AAAI format. The deadline for submissions is November 25, 2024 (AOE). The review process is double-blind, so submissions should be anonymized.
Submission Site Information: submissions will be managed through OpenReview: https://openreview.net/group?id=AAAI.org/2025/Bridge/Continual_Causality. New this year: we also host a challenge and allow for presentations of recently published works. These follow a separate call for participation at https://www.continualcausality.org
Bridge Committee
Keiland Cooper – University of California – kwcooper@uci.edu
Rebecca Herman – TU Dresden – rebecca.herman@tu-dresden.de
Martin Mundt – University of Bremen & TU Darmstadt – mundtm@uni-bremen.de
P. K. Srijith – IIT Hyderabad – srijith@cse.iith.ac.in
Devendra Singh Dhami – TU Eindhoven – d.s.dhami@tue.nl
Roshni Kamath – TU Darmstadt & hessian.AI – roshni.kamath@tu-darmstadt.de
Florian Busch – TU Darmstadt & hessian.AI – florian_peter.busch@tu-darmstadt.de
Bridge External URL
B9: Explainable AI, Energy and Critical Infrastructure Systems
The goal of this bridge will be to bring together researchers, practitioners from industry and policymakers to share technical advances and insights on applications and challenges for the application of AI in energy and critical infrastructure systems. The target audience includes AI researchers that are actively working on the use of AI and XAI in energy and critical infrastructure systems, as well as researchers in energy and critical infrastructure systems that would like to explore the potential benefits of applying AI and XAI to these domains. Other stakeholders, including policymakers and regulators interested in XAI, energy and critical infrastructure systems are welcome to join.
Format of Bridge
This 1-day bridge will include invited talks, panels and tutorials covering a wide range of topics at the intersection of XAI, Energy and Critical Infrastructure Systems. This bridge will also include networking sessions and mentoring opportunities for students.
Attendance
No fixed criteria to participate.
Submission Requirements
This Bridge accepts submissions to a variety of Tracks, which are described in the Bridge website. Our aim is to maximize opportunities and options for contributing to this Bridge and the Bridge community. Submissions to multiple Tracks are allowed.
Submissions need to discuss interactions between XAI, Energy and Critical Infrastructure systems in some fashion. All submissions should keep in mind that Bridge attendees are not necessarily experts in both fields. Presentations should try to bridge potential gaps, and welcome questions and discussion.
Submissions should be prepared using the AAAI 2025 template available here. Submissions should specify the Track they are submitted to, contain the names, affiliations and contact emails of the authors, and indicate which authors would be expected to attend the Bridge.
The full list of Track options is available at the bridge website along with full submission requirements and instructions and other important information. Submissions will be through EasyChair. The deadline for submitting to this bridge is November 25th, 2024. All accepted contributions must be presented in person. All participants, whether they have an accepted contribution or not, will be required to register to AAAI 2025 using the Bridge, Tutorial and Lab only registration.
All accepted submissions will be posted on this website. If a submission is accepted for physical presentation, authors will then be expected to provide any slides used in the presentation for inclusion in the Bridge website as well.
Submission Site Information: https://easychair.org/my/conference?conf=xaieci2025
Bridge Chair:
Francesco Leofante, Imperial College London, f.leofante@imperial.ac.uk
Bridge Committee:
Francesco Leofante, Imperial College London, f.leofante@imperial.ac.uk
André Artelt, Bielefeld University, aartelt@techfak.uni-bielefeld.de
Demetrios Eliades, University of Cyprus, eliades.demetrios@ucy.ac.cy
Anna Korre, Imperial College London, a.korre@imperial.ac.uk
Tim Miller, The University of Queensland, timothy.miller@uq.edu.au
Francesca Toni, Imperial College London, f.toni@imperial.ac.uk
Bridge External URL: https://www.doc.ic.ac.uk/~fleofant/aaai25-xai-eci/index.html
B10: Knowledge-guided Machine Learning: Bridging Scientific Knowledge and AI
Description
Knowledge-guided machine learning (KGML) is an emerging field of research that focuses on integrating scientific knowledge in ML frameworks to produce solutions that are scientifically grounded, explainable, and likely to generalize on out-of-distribution samples, even with limited training data. By using both scientific knowledge and data as complementary sources of introduction in the design, training, and evaluation of ML models, KGML seeks a distinct departure from black-box data-only methods and holds great potential for accelerating scientific discovery in a number of disciplines. The goal of our bridge is to nurture the cross-disciplinary community of researchers working at the intersection of AI and science by providing a common platform to catalyze and cross-fertilize ideas from diverse fields and shape the vision of the rapidly growing field of KGML.
Topics
We encourage participation on a range of topics exploring the synergy between scientific knowledge and ML, including (but not limited to): (a) use of scientific knowledge as loss functions or hard constraints in the training of ML models for supervised, unsupervised, and semi-supervised applications, (b) design of deep learning architectures that are grounded in scientific theories and generate explainable and physically meaningful feature representations, (c) use of simulated data generated by science-based models along with observations in ML frameworks, (d) techniques to augment imperfections or infer parameters in science-based models using ML, and (e) use of scientific knowledge in the design, pretraining, or finetuning of Foundation models in science.
Format
Our two-day bridge will include a mix of activities to support education, collaboration, and outreach in the field of KGML, including invited talks, lecture-style tutorials, hands-on demos, panel discussions, poster sessions, and networking/mentoring events.
Submission Requirements
We are accepting short submissions (maximum 2 pages excluding references) as extended abstracts or proposals in a variety of tracks such as: (a) Blue Sky Ideas papers providing a position or perspective of a research area in KGML, (b) Tutorials (lecture-style or hands-on demos) of KGML topics, (c) Posters showing preliminary results on cutting-edge research problems, (d) Datasets and Benchmarks papers relevant to KGML, (e) early career lightning talks promoting next-generation leaders in KGML including postdocs and early career investigators, and (f) dissertation forum submissions for graduate students to present their dissertation research in KGML. All submissions will undergo light review by the organizers for suitability for the bridge.
Submission Site Information: https://easychair.org/my/conference?conf=kgmlbridgeaaai25
Organizing Committee
- Arka Daw (dawa@ornl.gov)
- Nikhil Muralidhar (nmurali1@stevens.edu)
- Taniya Kapoor (t.kapoor@tudelft.nl)
- Kai-Hendrik Cohrs (kai.cohrs@uv.es)
Steering Committee
- Anuj Karpatne (karpatne@vt.edu)
- Xiaowei Jia (xiaowei@pitt.edu)
- Ramakrishnan Kannan (kannanr@ornl.gov)
- Vipin Kumar (kumar001@umn.edu)
Bridge URL: https://sites.google.com/vt.edu/kgml-bridge-aaai-25/
B11: Learning for Integrated Task and Motion Planning
Description:
Robotic agents are required to accomplish increasingly complex and longer-horizon tasks autonomously. This requires developing novel approaches for computing increasingly elaborate and robust plans that optimize the agents’ behavior and allow them to deal with unexpected events.
Effective solution approaches for such settings need to manage a rich coupling between three levels of abstraction – task, motion, and control. However, effectively integrating these three components has been established as a challenging sequential decision-making problem that requires integrating skills and tools from different research disciplines, investigated by different research communities which makes the integration of motions and high-level actions especially challenging.
Our proposed bridge program aims to bring together researchers from different research communities and help catalyze the next generation of research in combining AI, machine learning, and robotics and developing robots that are capable and efficient at all levels of deliberation and decision-making
Topics
Our bridge program will offer challenge problems, tutorials, laser talks, and panels on major elements of TMP and learning for TMP that are required to develop capable and dexterous autonomous robotic systems. The content will center around various related themes including motion planning, task planning, robust execution and control, perception, and manipulation, planning under uncertainty and risk, imitation and reinforcement learning, and more.
Format
We propose a two-day program. Each day will start with two lectures given by prominent researchers, including speakers who perform research at the crossover between learning, robotics, and AI. It will also include vision and challenge talks, to set the context for the field.
Because lectures alone are not enough to have a lasting impact, we will complement them with two hands-on lab sessions each day. During the labs, participants will implement ideas discussed in the talks and will solve problems in a simulated robotic setting.
To encourage the exchange of ideas, at the end of each lab session, we will facilitate a discussion on challenges that were encountered during the labs and potential solutions approaches. To further foster discussions, participants will be asked to provide 1-2 minute laser talks and will be encouraged to bring posters, which will be available throughout the day, with dedicated poster sessions during the breaks.
Attendance
The intended audience includes graduate students, postdocs, and researchers interested in developing capable real-world robotic systems, enabled through task and motion planning systems that combine a balance of model-based and machine learning methods. Of particular relevance are researchers from machine learning, perception, AI, Robotics, and control who are interested in this enterprise.
Bridge Committee
Sarah Keren, Technion – Israel Institute of Technology – sarahk@cs.technion.ac.il
Brian Williams, Massachusetts Institute of Technology – williams@csail.mit.edu
Michael Posa, University of Pennsylvania – posa@seas.upenn.edu
Bridge URL: https://github.com/CLAIR-LAB-TECHNION/AAAI_25_Bridge_TMP