How to Automatically Tune Machine Learning Settings Using Smart Math
A system that uses advanced statistical modeling to automatically find the best settings for complex machine learning models, saving engineers from manual trial and error.
Original patent title: “Systems and methods for Bayesian optimization using non-linear mapping of input”
A system that uses advanced statistical modeling to automatically find the best settings for complex machine learning models, saving engineers from manual trial and error. Granted to University of Toronto in 2022 with 24 claims.
Key facts
Coverage
What does this patent actually cover?
This patent describes a way to automate the tuning of machine learning systems, specifically by finding the best hyper-parameters. Hyper-parameters are the configuration settings, like learning rates or layer sizes, that define how a model learns. The system uses a probabilistic model—a mathematical way of representing uncertainty—to predict which settings will yield the best performance. The core innovation is applying a non-linear mapping to these settings before the model processes them, which helps the system better understand complex, non-obvious relationships between settings and outcomes. By iteratively evaluating the system and updating the model, it converges on the optimal configuration.
The gap
What does this patent NOT cover?
- Does not cover manual tuning of machine learning models by human engineers.
- Does not cover optimization methods that rely solely on linear transformations.
- Does not cover the actual architecture of the neural network being tuned, only the process of selecting its hyper-parameters.
- Does not cover non-probabilistic search methods like simple grid search or random search.
These exclusions are unique to PatentBrief — derived from the actual claim language, not patent-office boilerplate.
What made this novel
By using a non-linear one-to-one mapping to transform the input domain, the system can model complex, warped landscapes of hyper-parameters that traditional Gaussian processes might struggle to represent accurately.
Schematic visualization of the patent's claim structure. Hand-drawn diagrams in progress for each landmark patent.
Where you've seen this
Real-world examples
AutoML platforms for neural network architecture search
Hyper-parameter optimization frameworks for deep learning
Automated model training pipelines in cloud AI services
Why it matters
The bigger picture
Tuning machine learning models is notoriously time-consuming and expensive, often requiring massive computing resources. By automating this process, this technology enables faster development cycles and better-performing AI models. It represents a shift toward 'AutoML,' where the infrastructure itself learns how to optimize its own configuration.
Filed
September 4, 2018
Granted
November 15, 2022
Market context
Who's building on this
Companies in this space
The inventors are associated with the University of Toronto, a hub for deep learning research. Major cloud providers like Google, AWS, and Microsoft are actively building and deploying AutoML services that rely on similar Bayesian optimization techniques to streamline model development for their enterprise clients.
Market impact
This technology contributes to the broader industry move toward AutoML, which reduces the barrier to entry for deploying sophisticated AI. It helps companies reduce the 'compute tax' associated with model training by finding optimal configurations in fewer iterations, directly impacting the efficiency of large-scale machine learning operations.
Claim 1 — Plain English
What this patent covers
This patent describes a way to automate the tuning of machine learning systems, specifically by finding the best hyper-parameters. Hyper-parameters are the configuration settings, like learning rates or layer sizes, that define how a model learns. The system uses a probabilistic model—a mathematical way of representing uncertainty—to predict which settings will yield the best performance. The core innovation is applying a non-linear mapping to these settings before the model processes them, which helps the system better understand complex, non-obvious relationships between settings and outcomes. By iteratively evaluating the system and updating the model, it converges on the optimal configuration.
The clever bit
By using a non-linear one-to-one mapping to transform the input domain, the system can model complex, warped landscapes of hyper-parameters that traditional Gaussian processes might struggle to represent accurately.
What it does not cover
- Does not cover manual tuning of machine learning models by human engineers.
- Does not cover optimization methods that rely solely on linear transformations.
- Does not cover the actual architecture of the neural network being tuned, only the process of selecting its hyper-parameters.
- Does not cover non-probabilistic search methods like simple grid search or random search.
Patent timeline
Application submitted to the patent office
Application published, typically 18 months after filing
Patent officially issued
PatentBrief Score
Impact Score
Moderate
Citation count
0/40
No citations yet
Claim breadth
16/20
Broad claimsclaimsThe numbered statements at the end of a patent that legally define what the inventor owns.Read more →
Recency
20/20
Granted within 5 years
Assignee scale
20/20
Major company or institution
PatentBrief Impact Score — based on citation count, claim breadth, recency, and assignee scale. Not a legal assessment.
Heuristic Value Estimate
What this patent might be worth
$37K – $120K
Midpoint $75K · 12.2 yr remaining · industry ×1.6
Heuristic only — blends forward/backward citation counts, claim scope, time remaining, litigation history, and CPC-derived industry baseline. Real valuations need a professional appraisal.
The original legal language
Original claims
24 claims as filed with the patent office.
Concepts involved
Citations
Patent lineage
Cite this patent
Zemel, R., Swersky, K., Snoek, R. J., & Adams, R. P. (2022). How to Automatically Tune Machine Learning Settings Using Smart Math (U.S. Patent No. 11,501,192). U.S. Patent and Trademark Office. https://patentbrief.org/patent/us/11501192/alphafold-protein-structure-prediction
Auto-generated from the patent record. Double-check author order and the issue date against the official USPTO document before submitting.
Embed
Add this patent to your site
Drop this plain-English patent card into any blog post or article — free, no signup. It always links back to the full breakdown here.
<div data-patentlens-widget data-patent-number="US11501192"></div> <script src="https://patentbrief.org/embed.js" async></script>
Stay in the loop
Get a weekly digest of new patents.
One email per week. No spam. Unsubscribe anytime.
Keep exploring
Related patents you should know
US 4683195 · 1987
How to Make Billions of Copies of a DNA Segment
This patent describes the Polymerase Chain Reaction (PCR), a method to rapidly create many copies of a specific piece of DNA or RNA, enabling its detection and analysis.
Cetus Corp
US 8697359 · 2014
How to Edit Genes in Human Cells Using an Engineered CRISPR System
This patent describes an engineered CRISPR-Cas9 system for precisely cutting DNA in eukaryotic cells to change how genes work, opening the door for gene editing in complex organisms.
Massachusetts Institute of Technology
US 7657849 · 2010
How the iPhone's Slide-to-Unlock Gesture Works
Apple's 2010 patent describes unlocking a device by dragging a specific graphical image across the touchscreen along a predefined path, a gesture that became iconic with the original iPhone.
Apple Inc
US 4733665 · 1988
How Doctors Implant a Permanent Stent Using a Balloon
This patent describes the method for placing a permanent, expandable wire mesh tube inside a blood vessel or other body tube using a balloon-tipped catheter to widen it and keep it open.
Expandable Grafts Partnership
US 4965188 · 1990
How to Make Many Copies of a DNA Piece with Heat
This patent describes the Polymerase Chain Reaction (PCR) method, a technique to make millions of copies of a specific DNA segment using a heat-resistant enzyme and repeated temperature changes.
Cetus Corp
US 4235871 · 1980
How to Encapsulate Active Materials in Lipid Bubbles Efficiently
This patent describes a method for trapping biologically active substances inside tiny, multi-layered fat bubbles called liposomes, using a specific water-in-oil emulsion and gel-forming process to improve how much material gets captured.
Individual
More to explore
More in AI & Machine Learning
US 10452978 · 2019 · Google LLC
How AI Models Understand Language Using 'Attention'
US 6523026 · 2003 · Huntsman International LLC
How Computers Find Hidden Connections Between Different Fields of Knowledge
US 11615208 · 2023 · Capital One Services LLC
How Cloud Systems Automatically Create and Train AI Data Models
US 10402750 · 2019 · Facebook Inc
How Facebook Uses Deep Learning to Predict What You Might Like
New to patents?
Common Questions
Frequently Asked Questions
What does How to Automatically Tune Machine Learning Settings Using Smart Math cover?
A system that uses advanced statistical modeling to automatically find the best settings for complex machine learning models, saving engineers from manual trial and error.
Who owns patent US 11501192?
University of Toronto owns this patent, granted in 2022.
When does this patent expire?
This patent is expected to expire on November 15, 2042, when the invention enters the public domain.
What problem does this patent solve?
Tuning machine learning models is notoriously time-consuming and expensive, often requiring massive computing resources. By automating this process, this technology enables faster development cycles and better-performing AI models. It represents a shift toward 'AutoML,' where the infrastructure itself learns how to optimize its own configuration.
What does this patent NOT cover?
Does not cover manual tuning of machine learning models by human engineers.
Patent monitoring