Research initiative
Controllable Biology
Machine learning, causality, and control for intervention in biological systems
A new space for work at the intersection of learning, causality, and control, aimed at interventions in complex biological systems.
About
Controllable Biology is an academic initiative focused on modeling, identifying, and controlling biological systems with rigor and practical relevance.
Motivation
Biological systems exhibit rich, multi-scale dynamics. Making interventions trustworthy requires methods that integrate predictive modeling with causal and mechanistic structure.
Approach
We emphasize transparent assumptions, robust evaluation, and an end-to-end view of intervention pipelines, from data and models to policies and experimental feedback.
Community
The initiative supports workshops, curated talks, and collaboration across machine learning, statistics, control theory, and the life sciences.
Themes
Representative directions that connect learning and decision-making for biological intervention.
- Causal discovery and identifiability for biological networks and dynamical systems
- Control and reinforcement learning with safety, uncertainty, and constraints
- Mechanistic and hybrid modeling combining structure with scalable ML
- Experimental design and adaptive interventions with feedback loops
- Evaluation: counterfactual validation, robustness, and generalization
Organizers
Organizing team information can be listed here. Replace placeholders with names, affiliations, and links.
Name Surname
Name Surname
Speakers
Placeholder for confirmed and invited speakers. Add names, affiliations, and talk titles when available.
Coming soon. Speaker list and details will be posted here.
Workshop / Events
Placeholder for upcoming workshops, seminars, or related events. Include date, location, and registration details.
Coming soon. Event announcements and schedules will appear here.
Contact
For inquiries, collaboration proposals, or speaking interest, please reach out.
Updates
Add a newsletter or mailing list link here.
Replace placeholder contact information with your preferred addresses and links.