Mu Lab
Click to visit website
About
Mu Lab is an academic research group at Queen's University dedicated to advancing the fields of AI planning, multi-agent reasoning, and model understanding. The lab's primary mission is to develop techniques that allow AI systems to better comprehend complex environments and interact more effectively within multi-agent frameworks. By focusing on modeling uncertainty and interpreting model behavior, the team provides a foundation for more robust and transparent autonomous systems. Their work bridges the gap between theoretical AI research and practical application through a variety of specialized projects and peer-reviewed publications. The lab's technical contributions include a wide range of open-source tools and methodologies. For example, the PR2 project (PRP Rebooted) advances the state of the art in Fully Observable Non-Deterministic (FOND) planning, while the MACQ library implements action model acquisition techniques like LOCM. Other significant work involves translating business workflows (BPMN) into AI planning languages (PDDL) and leveraging Large Language Models (LLMs) to repair ill-defined models or incorrect plans through their "FixMyPlan" initiative. These tools are often shared via repositories like GitHub, allowing the broader AI community to test and build upon their findings. Mu Lab is primarily geared toward academic researchers, PhD students, and industry professionals in sectors like banking or healthcare who require sophisticated AI modeling solutions. Because their work often involves real-world case studies—such as detecting life events in bank conversations or predicting chemotherapy responses using radiomic signatures—it is highly relevant for organizations looking to apply automated planning and reasoning to specific domain challenges. The lab provides a collaborative environment where students and faculty work together to solve high-impact problems in automated reasoning. What distinguishes Mu Lab from other AI research entities is its deep focus on the intersection of classical AI planning and modern machine learning techniques. While many labs focus solely on deep learning, Mu Lab emphasizes model interpretability and the formal verification of plans. By integrating LLMs to assist in fixing plan errors and translating legacy business processes into formal AI models, they offer unique bridges between symbolic AI and contemporary data-driven approaches.
Pros & Cons
Offers open-source code for projects like the BPMN to PDDL translator and PR2 planner.
Conducts award-winning research, such as the Best Industry Paper at CASCON 2025.
Covers diverse applications ranging from banking conversation analysis to medical CT scan predictions.
Provides integrated tools that bridge Large Language Models with classical AI planning.
Maintains an active publication record in top-tier AI and medical informatics journals.
Primarily a research entity, so tools may require significant technical expertise to implement.
No dedicated commercial support or Service Level Agreements for the shared software.
Graduate positions and vacancies are currently listed as unavailable for new applicants.
Tools are research-oriented and may lack polished user interfaces for non-technical users.
Use Cases
Academic researchers can utilize the PR2 planner and MACQ library to benchmark new FOND planning algorithms.
Business analysts can translate complex BPMN workflows into PDDL format to automate process optimization using AI planning.
Data scientists in healthcare can apply the lab's radiomic signature research to predict patient responses to chemotherapy from CT data.
Software developers can implement continuous authentication systems by leveraging the lab's research on typing gait recognition.
AI engineers can use the FixMyPlan methodology to leverage LLMs for correcting errors in ill-defined domain models.
Platform
Task
Features
• continuous authentication
• action model acquisition
• multi-agent reasoning
• llm-based plan fixing
• radiomic signature prediction
• model uncertainty analysis
• bpmn to pddl translation
• fond planning (pr2)
FAQs
What kind of research does Mu Lab focus on?
The lab focuses on three core pillars: Model Understanding, Multi-agent Reasoning, and Modeling Uncertainty. They apply these concepts to AI planning, LLM integration, and real-world domain problems in finance and medicine.
Are the tools developed by Mu Lab available for public use?
Yes, many of their projects, such as the BPMN to PDDL translator and the MACQ library, include links to open-source code repositories. These are typically hosted on GitHub for community and academic access.
Can industry partners collaborate with the lab?
Yes, the lab has a history of industry collaboration, evidenced by their CASCON Best Industry Paper award for work on bank conversations. Interested parties should contact Prof. Muise directly via the provided email.
How can I access the lab's publications?
Most publications listed on the website include links to technical details and PDF downloads. Many are also linked to their corresponding open-source code for reproducibility.
Pricing Plans
Open Source
Free Plan• Access to research papers
• Open-source code repositories
• BPMN to PDDL translation tools
• PR2 planner resources
• MACQ library access
• Publicly available datasets
• Academic collaboration opportunities
Job Opportunities
There are currently no job postings for this AI tool.
Ratings & Reviews
No ratings available yet. Be the first to rate this tool!
Alternatives
Epoch AI
Track the trajectory of artificial intelligence through rigorous data on compute scaling, hardware trends, and model benchmarks for informed decision-making.
View DetailsPlanetarium
Improve the accuracy of AI planning by evaluating LLMs with a rigorous benchmark that uses a novel PDDL equivalence algorithm and 145,000+ text-to-PDDL pairs.
View Details
Digital Greenhouse
AI research lab developing cutting-edge AI solutions across multiple sectors.
View Details
Coastline Automation
Apply cutting-edge deep learning research to real-world projects like autonomous vehicles and image classification through detailed guides and practical case studies.
View Details
CyberNative.AI
Collaborate with interactive AI agents and researchers to prototype, test, and ship technology projects within an open, ethical social network for public good.
View Details
Essential AI
Build advanced software and STEM applications using frontier open-source language models, optimized datasets, and research-backed tools for developers and scientists.
View Details
Ai2
Access open-source foundational models and AI research to solve complex scientific, environmental, and robotic challenges through transparent development.
View Detailsyouare.ai
Discover dedicated algorithms and sources to enhance AI research and applications through personalized solutions.
View Details
Vianney.ai
Vianney.ai is the academic website of Professor Vianney Perchet, featuring his research in machine learning and game theory, publications, projects, and teaching materials.
View DetailsJohannes Hoffart
Johannes Hoffart is an AI researcher and CTO of SAP's AI Unit, focusing on Generative AI and Foundation Models on linked business data, with a strong publication record.
View Details
Competing in the Age of AI
Competing in the Age of AI explores how organizations are reinventing themselves using data, analytics, and AI to overcome traditional business constraints and drive growth.
View DetailsAI Breakthrough Hub
The AI Breakthrough Hub is a German initiative to advance AI research, development, and application, attracting top talent and fostering innovation.
View Details
TransferLab
Bridge the gap between machine learning research and industrial application with practical software, specialized trainings, and distilled technical AI knowledge.
View Details
GOAT.AI
Access cutting-edge research on Small Language Models (SLMs) and context length optimization to solve common AI issues like hallucinations and mode collapse.
View Details
SynthLabs
Empower AI research teams to scale reasoning and alignment using post-training technologies like Meta-CoT and generative reward models for safer foundation models.
View Details
Oscar AI
Accelerate healthcare operations and member support using AI-powered agents for medical record extraction, contract auditing, and personalized health guidance.
View Details
Research Center Trustworthy Data Science and Security
Enable safer digital systems through interdisciplinary research in machine learning, cybersecurity, and psychology to build human-centered AI technology.
View Details
LG AI Research
Accelerate industrial breakthroughs and scientific discovery using high-performance foundation models designed for language, materials, and biological intelligence.
View Details
IRT-AI
Develop robust, ethical, and trustworthy enterprise models through neurosymbolic research and knowledge-infused learning for mission-critical AI applications.
View DetailsFeatured Tools
adly.news
Connect with engaged niche audiences or monetize your subscriber base through an automated marketplace featuring verified metrics and secure Stripe payments.
View Details
Atoms
Launch full-stack products and acquire customers in minutes using a coordinated team of AI agents that handle everything from deep research to SEO and coding.
View Details
Sketch To
Convert images into artistic sketches or transform hand-drawn drafts into realistic photos using advanced AI models designed for artists, designers, and hobbyists.
View Details
Seedance 4.0
Create high-definition AI videos from text prompts or images in seconds with built-in audio, commercial rights, and support for multiple cinematic models.
View Details
Seedance
Transform text prompts or static images into cinematic 1080p videos with fluid motion and consistent multi-shot storytelling for creators and brands.
View Details
GenMix
Generate professional-quality AI videos, images, and voiceovers using world-class models like Sora 2 and Kling 2.6 through a single, unified creative dashboard.
View Details