Theoretical computer science·OpenAI Math·Result 114·Lean formalisation listed
Approximate counting of common integer polymatroid bases
One of the 372 results in OpenAI's public mathematics catalogue, filed under theoretical computer science: 2 preprints, the latest dated 5 October 2026. This page quotes OpenAI's own summary of the claim, links the papers and abstracts, and embeds the animation.
In one line: OpenAI reports a result titled ‘Approximate counting of common integer polymatroid bases’ (theoretical computer science) in 2 preprints, the latest dated 5 October 2026. It has not been peer reviewed. The channel's animation, Every OpenAI Math Result, Part 4 of 4: L = BPL, Integer Multiplication and 82 More, is on YouTube.
Chapter 11 of 84 in Every OpenAI Math Result, Part 4 of 4: L = BPL, Integer Multiplication and 82 More · starts at 3:28 · watch from there on YouTube →
What OpenAI claims
Gives a fully polynomial randomized approximation scheme for counting common integer bases of two integral polymatroids of equal total rank, supplied by exact rank-value oracles. Capacities are binary-encoded, each integer vector counts once, and oracle calls and bit operations outside the oracles are polynomial on every execution. For matroids presented by independence oracles, the results also cover common independent sets of prescribed, unrestricted, or maximum cardinality, even when the ranks differ. (Lean)
Papers (2)
An FPRAS for Common Integer Polymatroid Bases with Binary Capacities
OpenAI's abstract We give a fully polynomial randomized approximation scheme for counting common integer bases of two polymatroids with the same total rank, supplied by exact rank-value oracles. The total rank and capacities are encoded in binary, and each integer vector is counted once. On every execution, the number of oracle calls and the bit work outside the oracles are bounded by a fixed polynomial in the ground-set size, the binary input length, the inverse relative-error tolerance, and the logarithm of the inverse failure probability. The algorithm handles binary capacities directly, without expanding them into labelled copies.
Approximate counting of common bases of two matroids
OpenAI's abstract We give a fully polynomial randomized approximation scheme for counting the common bases of two arbitrary matroids of the same rank, supplied by independence oracles. The algorithm requires no explicit representation of either matroid and has polynomial bounds on both oracle calls and bit operations on every execution.
Status
Preprint. The papers have not been independently reviewed. Treat the result as OpenAI's claim until other mathematicians have checked it. If a video exists, it illustrates the statement; it is not the proof.
Quick answers
- Is it proven?
- Not independently yet. OpenAI posted 2 preprints, the latest dated 5 October 2026; it has not been peer reviewed, so treat the result as OpenAI's claim until other mathematicians have checked it.
- What exactly does OpenAI claim?
- Gives a fully polynomial randomized approximation scheme for counting common integer bases of two integral polymatroids of equal total rank, supplied by exact rank-value oracles. Capacities are binary-encoded, each integer vector counts once, and oracle calls and bit operations outside the oracles are polynomial on every execution. For matroids presented by independence oracles, the results also cover common independent sets of prescribed, unrestricted, or maximum cardinality, even when the ranks differ. (Lean)
- Where is the proof?
- In the paper, An FPRAS for Common Integer Polymatroid Bases with Binary Capacities (PDF).
- Is there a video?
- Yes, on YouTube (29:10, chapter 11).
Also in
- Chapter 9 of 9 in L = BPL, Matrix Multiplication Exponent, Integer Multiplication and 6 More · starts at 2:45 · watch from there on YouTube →
The video
Every OpenAI Math Result, Part 4 of 4: L = BPL, Integer Multiplication and 82 More · Compilation · 29:10 · chapter 11 from 3:28 · description, sources and chapters →
Related results in Theoretical computer science
- 115Sampling and counting contingency tables with arbitrary marginsVideoLean
- 116Uniform black-box noncommutative identity testing across characteristicsVideoLean
- 117Uniform sparsest cut: hardness and semidefinite gapsVideoLean
- 118Bin packing and unbounded configuration-LP gapsVideoLean
- 119The Courtade–Kumar and Hellinger conjecturesVideoLean
- 120Almost-linear-time exact matching and prescribed-degree factors in general graphsVideo