17963809. METHOD AND SYSTEM FOR EXTENDING QUERY PROCESSING WITH DIFFERENTIABLE OPERATORS simplified abstract (Microsoft Technology Licensing, LLC)

METHOD AND SYSTEM FOR EXTENDING QUERY PROCESSING WITH DIFFERENTIABLE OPERATORS

Organization Name

Microsoft Technology Licensing, LLC

Inventor(s)

Matteo Interlandi of Torrance CA (US)

Apurva Sandeep Gandhi of Union City CA (US)

Yuki Asada of Arlington MA (US)

Advitya Gemawat of Cambridge MA (US)

Victor Renjie Fu of Boston MA (US)

Lihao Zhang of Quincy MA (US)

Rathijit Sen of Redmond WA (US)

Dalitso Hansini Banda of Mountain View CA (US)

METHOD AND SYSTEM FOR EXTENDING QUERY PROCESSING WITH DIFFERENTIABLE OPERATORS - A simplified explanation of the abstract

This abstract first appeared for US patent application 17963809 titled 'METHOD AND SYSTEM FOR EXTENDING QUERY PROCESSING WITH DIFFERENTIABLE OPERATORS

Simplified Explanation

The patent application describes techniques for query processing over deep neural network runtimes.

• Receiving a query with a query operator and a trainable user-defined function (UDF)
• Determining a query representation based on the query
• Creating an initial neural network program with differentiable operators corresponding to the query operator
• Executing the neural network program in the runtime to generate a query result
• Training the initial neural network program to determine a trained neural network program
• Executing the trained neural network program in the runtime to generate inference information

Potential Applications

This technology could be applied in various fields such as natural language processing, image recognition, and recommendation systems.

Problems Solved

This technology helps in efficiently processing complex queries over deep neural network runtimes, improving performance and accuracy.

Benefits

The benefits of this technology include faster query processing, improved accuracy in results, and the ability to handle complex queries effectively.

Potential Commercial Applications

Potential commercial applications of this technology include search engines, e-commerce platforms, and data analytics companies.

Possible Prior Art

One possible prior art could be techniques for query processing in traditional database systems, which may not be as efficient or accurate as the methods described in this patent application.

What are the specific neural network runtimes used in this technology?

The specific neural network runtimes used in this technology are not explicitly mentioned in the abstract. Further details may be provided in the full patent application document.

How does this technology compare to existing query processing techniques in terms of efficiency and accuracy?

The abstract does not provide a direct comparison between this technology and existing query processing techniques. Further information on comparative studies or performance evaluations would be needed to answer this question accurately.

Original Abstract Submitted

Example aspects include techniques for query processing over deep neural network runtimes. These techniques include receiving a query including a query operator and a trainable user defined function (UDF). In addition, the techniques include determining a query representation based on the query, and determining, for performing the query in a neural network runtime, an initial neural network program based on the query representation, the initial neural network program including a differentiable operators corresponding to the query operator. and executing the neural network program in the neural network runtime over the neural network data structure to generate a query result. Further, the techniques include training the initial neural network program via the neural network runtime to determine a trained neural network program, and executing the trained neural network program in the neural network runtime to generate inference information.