17986001. RANKING MACHINE LEARNING PIPELINES USING JOINT COMPUTATIONS simplified abstract (INTERNATIONAL BUSINESS MACHINES CORPORATION)

RANKING MACHINE LEARNING PIPELINES USING JOINT COMPUTATIONS

Organization Name

INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor(s)

Dhavalkumar C. Patel of White Plains NY (US)

Srideepika Jayaraman of White Plains NY (US)

Shuxin Lin of White Plains NY (US)

Anuradha Bhamidipaty of Yorktown Heights NY (US)

Jayant R. Kalagnanam of Briarcliff Manor NY (US)

RANKING MACHINE LEARNING PIPELINES USING JOINT COMPUTATIONS - A simplified explanation of the abstract

This abstract first appeared for US patent application 17986001 titled 'RANKING MACHINE LEARNING PIPELINES USING JOINT COMPUTATIONS

Simplified Explanation

The patent application describes systems and methods for optimizing and training machine learning models through the use of ML pipelines and performance metrics.

• ML pipelines are executed using a subset of training data, with common data transformations implemented only once and shared between pipelines.
• Performance metrics are generated for each trained ML model based on validation data.
• Trained ML models are ranked based on performance metrics to generate a list of ranked models.
• The list of ranked ML models is outputted to a user.

Potential Applications

This technology can be applied in various fields such as finance, healthcare, marketing, and e-commerce for optimizing machine learning models and improving predictive accuracy.

Problems Solved

This technology solves the problem of efficiently training and optimizing machine learning models by sharing data transformations between pipelines and ranking models based on performance metrics.

Benefits

The benefits of this technology include improved model performance, reduced computational resources, and faster model training times.

Potential Commercial Applications

A potential commercial application of this technology could be in the development of personalized recommendation systems for e-commerce platforms, where accurate predictions are crucial for customer satisfaction and increased sales.

Possible Prior Art

One possible prior art for this technology could be the use of ensemble learning techniques to improve model performance by combining multiple models to make predictions.

How does this technology handle large-scale datasets and complex data structures?

The patent application does not provide details on how the technology handles large-scale datasets and complex data structures. Further information on scalability and adaptability to different data types would be beneficial.

What are the potential limitations or drawbacks of this technology?

The patent application does not discuss any potential limitations or drawbacks of the technology. Understanding any constraints or challenges associated with implementing this technology would be important for decision-making purposes.

Original Abstract Submitted

Systems and methods for optimizing and training machine learning (ML) models are provided. In embodiments, a computer implemented method includes: performing, by a processor set, a group execution of ML pipelines using a first subset of a training data set as input data for the ML pipelines, thereby generating a trained ML model for each of the ML pipelines, wherein data transformations that are common between the ML pipelines are implemented only once to generate an output, and the output is shared between the ML pipelines during the group execution of the ML pipelines; generating, by the processor set, performance metrics for each of the trained ML models based on validation data; ranking, by the processor set, the trained ML models based on the performance metrics, thereby generating a list of ranked ML models; and outputting, by the processor set, the list of ranked ML models to a user.