18272342. HIERARCHICAL ONLINE CONVEX OPTIMIZATION simplified abstract (Telefonaktiebolaget LM Ericsson (publ))
Contents
- 1 HIERARCHICAL ONLINE CONVEX OPTIMIZATION
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 HIERARCHICAL ONLINE CONVEX OPTIMIZATION - 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
HIERARCHICAL ONLINE CONVEX OPTIMIZATION
Organization Name
Telefonaktiebolaget LM Ericsson (publ)
Inventor(s)
Gary Boudreau of Kanata, Ontario (CA)
Hatem Abou-zeid of Calgary, Alberta (CA)
Juncheng Wang of Toronto, Ontario (CA)
Ben Liang of Whitby, Ontario (CA)
Min Dong of Whitby, Ontario (CA)
HIERARCHICAL ONLINE CONVEX OPTIMIZATION - A simplified explanation of the abstract
This abstract first appeared for US patent application 18272342 titled 'HIERARCHICAL ONLINE CONVEX OPTIMIZATION
Simplified Explanation
The method described in the abstract involves performing online convex optimization by receiving local decision vectors and data from multiple worker nodes, performing multi-step gradient descent, determining a global decision vector and global information, and sending this information back to the worker nodes.
- The method involves receiving local decision vectors and data from multiple worker nodes.
- Multi-step gradient descent is performed based on the received local decision vectors and data.
- A global decision vector and corresponding global information are determined.
- The global decision vector and global information are sent back to each worker node.
Potential Applications
The technology described in this patent application could be applied in various fields such as machine learning, data analysis, and optimization algorithms.
Problems Solved
This technology helps in efficiently optimizing convex functions in an online setting by utilizing multiple worker nodes to perform gradient descent and share information.
Benefits
The benefits of this technology include faster optimization of convex functions, improved scalability, and the ability to handle large datasets distributed across multiple nodes.
Potential Commercial Applications
Potential commercial applications of this technology include online advertising optimization, financial portfolio management, and real-time data analysis for large-scale systems.
Possible Prior Art
One possible prior art for this technology could be the use of distributed computing and parallel processing techniques in optimization algorithms.
What are the specific industries that could benefit from this technology?
Industries such as e-commerce, finance, and healthcare could benefit from this technology by optimizing their decision-making processes and improving the efficiency of their operations.
How does this technology compare to existing methods for online convex optimization?
This technology improves upon existing methods by utilizing multiple worker nodes to perform gradient descent in a distributed manner, allowing for faster and more efficient optimization of convex functions.
Original Abstract Submitted
A method for performing online convex optimization is provided. The method includes receiving, from two or more worker nodes, a local decision vector and local data corresponding to each of the two or more worker nodes. The method includes performing a multi-step gradient descent based on the local decision vector and the local data received from the two or more worker nodes. Performing the multi-step gradient descent includes determining a global decision vector and corresponding global information. The method includes sending, to each of the two or more worker nodes, the global decision vector and corresponding global information.
