Computation Graph Fusion for Fewer Table Lookup Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing homomorphic encryption techniques face significant overhead, particularly in table lookup operations, when applied to large amounts of data or complex computations, necessitating optimization for efficient execution.
Innovation Solution
A method and system that optimize computer programs by transforming the computation graph to fuse multiple nodes into a single table lookup operation, automatically determining which nodes can be implemented as such based on their connections in the graph, reducing the number of table lookup operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If homomorphic encryption techniques are applied to large amounts of data or complex computations, then data privacy and security are improved, but computational overhead and execution time increase significantly
Solution Approach 1:
The patent merges multiple sequential operations (addition, subtraction, scalar multiplication, and table lookup) into a single homomorphic table lookup operation. By combining these operations that were previously executed separately into one unified homomorphic operation, the patent reduces the number of cryptographic operations required, thereby decreasing computational overhead and execution time while maintaining data privacy protection through homomorphic encryption
2Adaptability or versatility
If multiple separate operations are performed in homomorphic encryption, then operational flexibility is maintained, but the number of table lookup operations increases
Solution Approach 1:
The patent combines multiple separate operations (addition, subtraction, scalar multiplication, and table lookup) into a single homomorphic table lookup operation. This merging reduces the total number of table lookup operations required while preserving the functional capabilities of the original separate operations, thus improving productivity without sacrificing operational flexibility
Solution Approach 2:
The patent implements a universal homomorphic table lookup operation that can perform multiple functions (addition, subtraction, scalar multiplication, and table lookup) within a single operation. This multi-functional approach allows the system to maintain operational flexibility for different computational needs while reducing the overall number of operations required, thereby improving productivity
Data Source
AI summary
The invention relates to a computer-implemented method (800) of optimizing a computer program for an execution environment that supports a table lookup operation. The computer program is represented as a computation graph, wherein respective nodes of the computation graph represent respective operations. For respective nodes, it is determined whether or not the respective node can be implemented by a table lookup operation. For a node that can be implemented by a table lookup operation, it is determined that one or more further nodes of the computation graph can be fused into the table lookup operation. A transformed representation of the computer program is output wherein the node and the one or more further nodes are fused into a single table lookup operation.