Yale FDS · Workshop 2026
Home About Schedule
Register →
Yale FDS Workshop · April 24, 2026

About the Workshop

Audience Use Cases Prerequisites Infrastructure Speakers Organizers
Intended Audience

Who is this for?

This workshop is for researchers interested in using computers to find, build, and discover things. It must be possible for a computer to know when it has succeeded and for it to know when it is close.

We encourage attendees from mathematical and physical sciences, and engineering. Graduate students, postdoctoral researchers, and faculty are all welcome. You do not need to be at Yale.

Use Cases

What can you do with ShinkaEvolve?

Here are some examples of the kinds of problems these tools can tackle. If your research involves anything like these, this workshop is for you.

√

Disproving mathematical conjectures

If you want to prove an inequality f(x) ≤ 0, you can set up a search for x* such that f(x*) > 0. Finding such a counterexample disproves the conjecture. ShinkaEvolve can search complex, high-dimensional spaces for these rare witnesses far more effectively than random sampling or conventional optimization.

▦

Discovering structures in combinatorics and complexity theory

LLM-guided search has been used to find graphs with unusual properties — large cuts, large independent sets — that serve as gadgets in hardness-of-approximation proofs. Brute force search becomes impractical even for small graphs; evolutionary search finds them.

○

Neural network architecture search

Given an initial model and an evaluation metric — say, validation accuracy on a domain-specific dataset — ShinkaEvolve can iteratively modify the architecture, adding layers, changing connectivity, and trying different training strategies to improve performance.

⬡

Molecular representation discovery

ShinkaEvolve can evolve algorithms that decompose molecular graphs into structural motifs, improving the accuracy of downstream models that predict properties like toxicity or biological activity.

∫

Numerical integrator design

Given an initial ODE solver, ShinkaEvolve can propose modifications to the integration strategy — alternative operator splittings, time discretizations, interpolation schemes — to reduce numerical error or improve stability.

⦵

Quantum circuit optimization

ShinkaEvolve can search over circuit templates which find the ground state energy of a quantum system or which realize desired operators, optimizing both the circuit structure and its parameters simultaneously.

…

Biology & bioscience coming soon

We are speaking with researchers in the biological sciences about relevant applications. Check back soon.

Prerequisites

What do you need to know?

Essential
  • A laptop with a command line environment and a text editor you are comfortable using
  • Basic familiarity with at least one programming language
  • Comfort reading documentation and code
Helpful
  • Experience connecting to compute clusters
  • Experience coding in Python
  • Experience using LLM coding agents such as Claude Code or Cursor
Not required
  • Prior experience formulating research problems as programmatic search
  • Familiarity with ShinkaEvolve or any other LLM-guided search tool
  • You will learn these on the day
Infrastructure

Everything ready on arrival

All software is pre-installed on Yale's Bouchet research computing cluster. Participants receive a temporary account — just SSH in and start working. No local setup required on the day. Or, bring your laptop and hack on things with your personal setup.

Compute

Temporary Bouchet accounts provisioned for all registered participants. ShinkaEvolve and Claude Code pre-installed and ready to use.

API Access

Each participant receives an individual API key with a spending limit, pre-configured in the cluster environment.

Claude Pro

Participants without an existing Claude Pro or Max subscription will receive a one-month gift subscription (~$20) prior to the event. You will be asked about your subscription status at registration.

On-site Support

Two YCRC staff members present for the full day to troubleshoot any technical issues.

Registration closes a few days before April 24 to allow time for account provisioning.

Further Reading

References

  1. Georgiev, B., Gómez-Serrano, J., Tao, T., & Wagner, A. Z. (2025). Mathematical exploration and discovery at scale. arXiv:2511.02864.
  2. Lange, R. T., Imajuku, Y., & Cetin, E. (2025). ShinkaEvolve: Towards open-ended and sample-efficient program evolution. arXiv:2509.19349.
  3. Novikov, A., Vũ, N., Eisenberger, M., et al. (2025). AlphaEvolve: A coding agent for scientific and algorithmic discovery. arXiv:2506.13131.
  4. Tao, T. (2025, May 15). Mathematical applications of LLM-guided evolutionary search [Post]. Mathstodon. mathstodon.xyz/@tao/114508029896631083
Invited Speakers

Researchers at the frontier

RL

Rob Lange

Sakana AI · Creator of ShinkaEvolve · via Zoom

Researcher at Sakana AI and lead developer of ShinkaEvolve. Rob will join the workshop remotely to give a keynote talk. His work focuses on open-ended evolutionary search and making these tools accessible to the research community.

AN

Alexander Novikov

Google DeepMind · Lead author, AlphaEvolve · attendance TBD

Lead author of AlphaEvolve. Alex’s research at Google DeepMind has shaped how the field thinks about AI-guided scientific discovery.

Organizers

Yale FDS

MP

Marco Pirazzini

PhD Student, Computer Science, Yale · Tutorial Lead
AC

Antares Chen

PhD Student, Computer Science, University of Chicago · Tutorial Lead
EH

Emily Hau

Associate Director, Yale Institute for Foundations of Data Science
DS

Dan Spielman

Sterling Professor of Computer Science and Statistics & Data Science, Yale

Opening session: Optimization for Design and Discovery

Yale Institute for Foundations of Data Science
Yale Center for Research Computing
Kline Tower · New Haven · 2026
Home · Schedule · Register
FDS Workshop: AI for Scientific Discovery
FDS Workshop: AI for Scientific Discovery
  • Home
  • About
  • Schedule
  • Register