Qualcomm incorporated (20240095513). FEDERATED LEARNING SURROGATION WITH TRUSTED SERVER simplified abstract
Contents
- 1 FEDERATED LEARNING SURROGATION WITH TRUSTED SERVER
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 FEDERATED LEARNING SURROGATION WITH TRUSTED SERVER - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
FEDERATED LEARNING SURROGATION WITH TRUSTED SERVER
Organization Name
Inventor(s)
Jian Shen of San Diego CA (US)
Jamie Menjay Lin of San Diego CA (US)
FEDERATED LEARNING SURROGATION WITH TRUSTED SERVER - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240095513 titled 'FEDERATED LEARNING SURROGATION WITH TRUSTED SERVER
Simplified Explanation
The present disclosure relates to techniques for surrogated federated learning, where a trusted server receives a set of intermediate activations from a node device, refines weights associated with a neural network using the activations, and transmits weight updates to a federated learning system.
- Intermediate activations received at a trusted server from a node device.
- Refinement of weights associated with a neural network using the set of intermediate activations.
- Transmission of weight updates to a federated learning system.
Potential Applications
The technology described in this patent application could be applied in various fields such as healthcare, finance, and autonomous vehicles for collaborative learning without sharing raw data.
Problems Solved
This technology solves the problem of privacy concerns associated with sharing sensitive data in federated learning systems by using surrogated intermediate activations instead.
Benefits
The benefits of this technology include improved privacy protection, enhanced collaboration in machine learning tasks, and increased efficiency in federated learning processes.
Potential Commercial Applications
A potential commercial application of this technology could be in the development of secure and efficient machine learning models for industries that require collaborative data analysis, such as healthcare and finance.
Possible Prior Art
One possible prior art in this field is the concept of federated learning, where models are trained across multiple decentralized devices without sharing raw data.
Unanswered Questions
How does this technology ensure the security of the transmitted weight updates?
The patent abstract does not provide details on the security measures implemented to protect the weight updates during transmission.
What are the computational requirements for implementing surrogated federated learning in real-world applications?
The abstract does not mention the computational resources needed to deploy this technology on a large scale.
Original Abstract Submitted
certain aspects of the present disclosure provide techniques and apparatus for surrogated federated learning. a set of intermediate activations is received at a trusted server from a node device, where the node device generated the set of intermediate activations using a first set of layers of a neural network. one or more weights associated with a second set of layers of the neural network are refined using the set of intermediate activations, and one or more weight updates corresponding to the refined one or more weights are transmitted to a federated learning system.
