Offline Reinforcement Learning Patents: Who Leads and Where the Field Is Heading
Concentrated at the top. The five leading assignees account for 175 of 285 records in scope (61.4%), and the leading ten hold 71.9%. Annual filings sat at 0 in 2017 and stayed low until 2021 (4 records). Growth from there was steep: 2021 to 2024 saw a 475% increase, reaching 23 records in 2024, with the dataset's peak year so far at 123 records in 2023. Figures for 2025 and 2026 are still filling in due to publication lag and should not be read as decline.
- 1STRONG FORCE VCN PORTFOLIO 2019 LLC96
- 2STRONG FORCE TX PORTFOLIO 2018 LLC44
- 3GDM HOLDING LLC15
- 4FIGURE AI INC12
- 5ANCESTRY COM OPERATIONS INC8
See the full offline reinforcement learning analysis in Eureka
- The complete ranking, not just the top five
- Every IPC branch with its share of the corpus
- The most-cited records, and where claim space is still thin
Common questions about offline reinforcement learning patents
What counts as an offline reinforcement learning patent?
In this dataset, a record qualifies if it combines offline-RL or behaviour-cloning terminology, such as distribution shift, with a conservative-optimization mechanism like value overestimation control, conservative regularization, dataset coverage, importance sampling, policy constraint, or replay buffer management. This is a technical, not a branding, definition: many qualifying filings are framed as robotics or recommendation-system patents rather than pure machine-learning patents. The 285 records in this scope were filed between 2015 and mid-2026.
Who are the leading filers of offline reinforcement learning patents?
Filing is concentrated at the top: the leading five assignees together hold 175 of 285 records in scope, or 61.4%, and the leading ten hold 71.9%. The single leading assignee alone holds 96 records, far ahead of the fifth-place assignee at 8. Below the top ten, the ranking spreads thinly across dozens of assignees with only one or two records each, which is typical of a field still in an active build-out phase.
Is offline reinforcement learning patent filing still growing in 2026?
Filings grew sharply from 4 in 2021 to 23 in 2024, a 475% increase, with the busiest year on record being 2023 at 123 filings. Figures shown for 2025 and 2026 are lower, but that reflects the roughly 18-month lag between filing and publication rather than a genuine slowdown; those years are still filling in. The safest complete year for growth comparisons is 2024.
Disclaimer. This analysis is based on Patsnap Eureka data drawn from a limited snapshot of global patent records and is provided for general information and reference only. Patent data carries inherent limitations — recent filings are under-counted because of publication lag, counts may be on a record or family basis, classification and applicant-name data may contain errors or duplicates, and the underlying search query defines the scope shown — so the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing here is an exhaustive prior-art, novelty, freedom-to-operate or validity search, nor does it constitute legal, financial or professional advice, and it should not be relied upon as such. Verify independently and review with qualified patent and legal professionals before acting on it.
Method: Filing trend and technology composition. Derived from a Patsnap search on Offline Reinforcement Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish. Every share divides by all records in scope. Data: Patsnap Eureka. See the full landscape report.