Polynomial Approximation for FHE Branching Operations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fully homomorphic encryption (FHE) systems cannot perform branching operations, such as 'if' statements, which are essential for many algorithms, leading to security risks when data is decrypted for processing.

Innovation Solution

Approximate branching operations using polynomial approximations, allowing computations to be performed directly on FHE encrypted data without decryption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If data is decrypted for processing to enable branching operations, then computational functionality is improved, but data security deteriorates

Engineering Contradiction:
Improvecomputational functionalityVSAvoiddata security
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces an intermediary mechanism (homomorphic encryption system with approximation algorithms) that enables branching operations to be performed on encrypted data without decryption. The encrypted data acts as an intermediary representation that preserves security while allowing computational functionality through polynomial approximation of branch operations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the parameter state of the data from decrypted plaintext to encrypted ciphertext, allowing branching operations to be performed through polynomial approximations. By operating in the encrypted domain with modified parameters (encrypted values instead of plaintext), the system achieves both security and computational functionality.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If polynomial approximation is used to enable branching operations on encrypted data, then data security is improved, but computational precision deteriorates

Engineering Contradiction:
Improvedata securityVSAvoidcomputational precision
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies partial approximation rather than exact representation of branching operations. By using polynomial approximations that are sufficiently accurate for the intended applications but not necessarily exact, the system achieves a practical balance between security preservation and computational precision.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent accepts a change in the precision parameter as a trade-off for achieving data security. The polynomial approximation introduces controlled imprecision, but this parameter change enables the critical security improvement by allowing operations on encrypted data without decryption.

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If branching operations are performed on encrypted data using polynomial approximation, then ease of operation is improved, but device complexity increases

Engineering Contradiction:
Improveease of operationVSAvoiddevice complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent extracts the branching operation capability from the decryption process. By taking out the ability to perform conditional logic from the plaintext domain and implementing it in the ciphertext domain through polynomial approximation, the system simplifies the operational flow (no decryption needed) while adding complexity to the underlying computational mechanism.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3888287B1System and method for approximating branching operations for use with data encrypted by fully homomorphic encryption (FHE)
Publication Date: 2025.10.22 INTUIT INC
  • EP3888287B1 patent drawingFigure 1
  • EP3888287B1 patent drawingFigure 2
  • EP3888287B1 patent drawingFigure 3

AI summary

Systems and methods that approximate and use branching operations on data encrypted by fully homomorphic encryption (FHE). The systems and methods may use polynomial approximation to convert "if" statements into "soft if" statements that may be applied to the FHE encrypted data in a manner that preserves the security of the systems and methods.