18278101. INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM simplified abstract (NEC Corporation)
Contents
- 1 INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM - 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 Unanswered Questions
- 1.11 Original Abstract Submitted
INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM
Organization Name
Inventor(s)
INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM - A simplified explanation of the abstract
This abstract first appeared for US patent application 18278101 titled 'INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM
Simplified Explanation
The abstract describes an inference device that receives an observation as input, generates hypothesis candidates, converts them into ILP or SAT problems, generates equivalent problems with variable order changes, solves them in parallel using multiple solvers, and outputs the optimal solution.
- Observation input means receives an observation as input
- Hypothesis candidate generation means generates hypothesis candidates using inference knowledge
- Problem conversion means converts hypothesis candidates into ILP or SAT problems
- Equivalent problem generation means creates equivalent ILP or SAT problems with variable order changes
- Solver parallelization means solves equivalent problems in parallel using multiple solvers
- Optimal solution output means outputs the result of the first solver to produce a solution
Potential Applications
This technology can be applied in various fields such as data analysis, pattern recognition, and decision-making processes.
Problems Solved
1. Efficiently generating and solving equivalent ILP or SAT problems 2. Optimizing the output of the solver by parallelizing the solving process
Benefits
1. Faster and more efficient problem-solving 2. Improved accuracy in generating optimal solutions
Potential Commercial Applications
Optimizing supply chain logistics, improving financial forecasting models, enhancing cybersecurity systems.
Possible Prior Art
Prior art may include similar inference devices used in machine learning algorithms or optimization techniques.
Unanswered Questions
How does this technology compare to existing parallel solving methods?
This article does not provide a direct comparison to existing parallel solving methods, leaving the reader to wonder about the specific advantages of this approach.
What are the limitations of this technology in terms of scalability?
The article does not address the scalability limitations of this technology, leaving open questions about its performance with larger and more complex problems.
Original Abstract Submitted
In an inference device, an observation input means receives an observation as an input. A hypothesis candidate generation means generates hypothesis candidates by applying inference knowledge to the observation in the backward direction. The problem conversion means converts the hypothesis candidate into an ILP problem or a SAT problem. An equivalent problem generation means generates a specified number of equivalent ILP problems or equivalent SAT problems in which an order of the variables included in the converted ILP problem or SAT problem is changed. The solver parallelization means solves the generated equivalent ILP problems or equivalent SAT problems by executing a specified number of identical ILP solvers or SAT solvers in parallel. The optimal solution output means outputs a result of the ILP solver or the SAT solver that output the result first, among the specified number of ILP solvers or SAT solvers, as the optimal solution.
