18166027. PRIVACY ENHANCED FEDERATED TRAINING AND INFERENCE OVER VERTICALLY AND HORIZONTALLY PARTITIONED DATA simplified abstract (International Business Machines Corporation)

PRIVACY ENHANCED FEDERATED TRAINING AND INFERENCE OVER VERTICALLY AND HORIZONTALLY PARTITIONED DATA

Organization Name

International Business Machines Corporation

Inventor(s)

Swanand Ravindra Kadhe of San Jose CA (US)

Heiko H. Ludwig of San Francisco CA (US)

Nathalie Baracaldo Angel of San Jose CA (US)

Yi Zhou of San Jose CA (US)

Alan Jonathan King of South Salem NY (US)

Keith Coleman Houck of Rye NY (US)

Ambrish Rawat of Dublin (IE)

Mark Purcell of Naas (IE)

Naoise Holohan of Maynooth (IE)

Mikio Takeuchi of Yokohama (JP)

Ryo Kawahara of Toshima-Ward (JP)

Nir Drucker of Zichron Yaakov (IL)

Hayim Shaul of Kfar Saba (IL)

PRIVACY ENHANCED FEDERATED TRAINING AND INFERENCE OVER VERTICALLY AND HORIZONTALLY PARTITIONED DATA - A simplified explanation of the abstract

This abstract first appeared for US patent application 18166027 titled 'PRIVACY ENHANCED FEDERATED TRAINING AND INFERENCE OVER VERTICALLY AND HORIZONTALLY PARTITIONED DATA

Simplified Explanation:

This patent application relates to systems, devices, and methods for federated training and inferencing. It involves training an inferential model using data from multiple parties without directly sharing private data.

• The system includes a memory storing computer executable components and a processor executing these components.
• The modeling component trains the inferential model using horizontally and vertically partitioned data with a random decision tree.
• The inference component generates inferences in response to queries without sharing first party private data.

Key Features and Innovation:

• Federated training and inferencing system
• Use of random decision tree for training inferential model
• Protection of first party private data during inferencing

Potential Applications:

This technology can be applied in various fields such as healthcare, finance, and marketing for collaborative data analysis without compromising data privacy.

Problems Solved:

• Secure inferencing without sharing sensitive data
• Efficient training of inferential models using distributed data sources

Benefits:

• Enhanced data privacy protection
• Improved collaboration among multiple parties
• Efficient and accurate inferencing process

Commercial Applications:

Potential commercial applications include secure data analysis platforms for industries that require collaborative data processing while maintaining data privacy.

Questions about Federated Training and Inferencing:

1. How does federated training differ from traditional centralized training methods? 2. What are the key challenges in implementing federated training and inferencing in real-world applications?

Frequently Updated Research:

Stay updated on the latest advancements in federated learning and data privacy regulations to ensure compliance and optimize performance.

Original Abstract Submitted

Systems, devices, computer program products and/or computer-implemented methods of use provided herein relate to federated training and inferencing. A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory, wherein the computer executable components can comprise a modeling component that trains an inferential model using data from a plurality of parties and comprising horizontally partitioned data and vertically partitioned data, wherein the modeling component employs a random decision tree comprising the data to train the inferential model, and an inference component that responds to a query, employing the inferential model, by generating an inference, wherein first party private data, of the data, originating from a first passive party of the plurality of parties, is not directly shared with other passive parties of the plurality of parties to generate the inference.