OpenAI Blog
Apr 5, 2018
Retro Contest
We’re launching a transfer learning contest that measures a reinforcement learning algorithm’s ability to generalize from previous experience.
Retro Contest: Results
The first run of our Retro Contest—exploring the development of algorithms that can generalize from previous experience—is now complete.
Related stories
MIT News AI
Jun 17
In game theory, generalists sometimes win out over specialists
Researchers show that for certain kinds of games, an overlooked class of algorithms performs much better than expected.
arXiv AI
Jul 14
GrandCode: Achieving Grandmaster Level in Competitive Programming via Agentic Reinforcement Learning
arXiv:2604. 02721v2 Announce Type: replace Abstract: Competitive programming remains one of the last few human strongholds in coding against AI.
arXiv Machine Learning
Sep 3
Towards Solving the Gilbert-Pollak Conjecture via Large Language Models
The paper announces a new lower bound of 0.8559 for the Steiner ratio, improving on the previous 0.824 bound for the Gilbert‑Pollak Conjecture. It introduces an AI system that uses large language models to generate rule‑constrained geometric lemmas, which are then turned into executable verification functions that certify the bound. The approach relies on only thousands of LLM calls, highlighting the feasibility of LLM‑based methods for advanced mathematical research.
arXiv Machine Learning
Jul 21
From Heuristic Selection to Automated Algorithm Design: LLMs Benefit from Strong Priors
arXiv:2603. 02792v2 Announce Type: replace Abstract: Large Language Models (LLMs) have already been widely adopted for automated algorithm design, demonstrating strong abilities in generating and evolving algorithms across various fields.
arXiv AI
Jun 17
First Proof Second Batch
arXiv:2606. 18119v1 Announce Type: new Abstract: To assess the ability of current AI systems to correctly solve research-level mathematics problems, we tested several AI systems on a set of ten problems in a broad range of mathematical fields; these problems arose naturally in the research process of the contributors.
arXiv AI
Sep 4
Counterfactual Routing Using Integer Programming with Constraint Generation
The paper "Counterfactual Routing Using Integer Programming with Constraint Generation" presents a solution to the IJCAI 2025 Counterfactual Routing Competition. The authors model the problem as an integer program and iteratively add constraints until an exact solution is found. In evaluation on held‑out test instances, their method ranked fourth in solution quality and was the fastest, averaging 9.0 seconds versus 118.8 seconds for the next‑fastest submission.
arXiv AI
Jun 29
SEA-TS: Self-Evolving Agent for Autonomous Code Generation of Time Series Forecasting Algorithms
arXiv:2603. 04873v3 Announce Type: replace Abstract: Accurate time series forecasting underpins decision-making in many domains, yetconventional ML development often faces data scarcity, distribution shift, anddiminishing returns from manual iteration.
OpenAI Blog
Nov 24, 2025
GPT-5 and the future of mathematical discovery
UCLA Professor Ernest Ryu and GPT-5 solved a key question in optimization theory, showcasing AI’s role in accelerating mathematical discovery.
arXiv AI
Jul 2
AlgoBench: Benchmarking Algorithmic Adaptation in Code Generation
arXiv:2607. 00062v1 Announce Type: cross Abstract: High pass rates on established programming benchmarks such as HumanEval and LiveCodeBench do not always show whether a model can reason about algorithms.
arXiv Machine Learning
Jun 15
The Program Is Still There: A Conservation Law for Program Discovery
arXiv:2606. 13799v1 Announce Type: cross Abstract: Finding the shortest program that generates a sequence is uncomputable, and for six decades that fact has been mistaken for a wall around finding any generating program.
arXiv Machine Learning
Jun 30
Learning Deterministic Finite-State Machines from the Prefixes of a Single String is NP-Complete
arXiv:2601. 12621v2 Announce Type: replace-cross Abstract: It is well known that computing a minimum deterministic finite automaton consistent with a given set of positive and negative examples is NP-hard.