# US 11,308,402 B2 · Low probability transitions and boundary crossing into disallowed states for a more optimal solution > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,308,402 B2 - Document ID: US11308402B2 - Status: Granted - Kind: B2 - Application number: 16/919,188 - Filed: 2020-07-02 - Published: 2020-10-22 - Granted: 2022-04-19 - Earliest claimed: 2017-10-30 - Assignees: The Aerospace Corporation - Applicants: The Aerospace Corporation - Inventors: Terence Yeoh; Nehal Desai - Web page: https://patentgrep.com/patent/US11308402B2 ## CPC - G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods} - G06F 18/2135: based on approximation criteria, e.g. principal component analysis - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 5/025: Extracting rules from data - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/00: Machine learning - G06N 5/02: Knowledge representation; Symbolic representation - G06V 10/40: Extraction of image or video features - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Artificial intelligence (AI) techniques that map disallowed states and enable access to those states under certain conditions through a search algorithm are disclosed. In other words, scenario boundaries may be crossed by jumping from one scenario that is less desirable or even has no solution to another scenario that is more desirable. ## Claims ### Claim 1 A computer-implemented method, comprising: by a computing system, performing principal component analysis (PCA) on raw data resulting in a UID comprising six numbers, with three of the six numbers resulting from principal components of a 2D wavelet transform of a respective image and another three of the six numbers resulting from principal components of a discrete cosine transform (DCT) of the respective image to generate relational sequences in trade space, wherein the trade space represents relative relationships in the raw data; mapping a solution space based on the generated relational sequences, by the computing system; replaying a plurality of scenarios, by the computing system, to determine pinch points where path structures are proximate to one another and more favorable outcomes can be jumped to; and generating a new solution path, by the computing system, based on one or more jumps to more favorable outcomes. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, further comprising: seeding the plurality of scenarios with initial conditions, by the computing system, wherein the initial conditions are seeded with real world data; and initializing random states, by the computing system, to explore more preferable solutions. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, further comprising: saving each relational sequence into a positive, negative, or neutral bin based on its outcome, by the computing system. ### Claim 4 (depends on 3) The computer-implemented method of claim 3, wherein the neutral bin is a default for observations of natural phenomena in order to simulate natural phenomena that cannot be changed. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein a frequency with which a path including a given node is chosen is correlated to a number of times that the node has appeared during past simulations. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein when a negative note is inevitable on all paths, a chance that a jump is attempted is increased. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the six numbers are between 1 and 100. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, wherein the UID further comprises numbers representing sentiment. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein the UID comprises two additional numbers that are used to facilitate pattern lock, and the pattern lock is utilized for cases where states are identical to enable repeating states to occur and still retain a sequence. ### Claim 10 (depends on 9) The computer-implemented method of claim 9, wherein upon choosing a next possible step, the method further comprises: evaluating prior experiences from the pattern lock, by the computing system; choosing a best experienced outcome, by the computing system; searching adjacent UIDs that have been previously experienced or simulated, by the computing system; and evaluating, by the computing system, whether a potential improvement of outcome warrants choosing a new path that has not been experienced. ### Claim 11 (depends on 1) The computer-implemented method of claim 1, wherein the UID comprises an additional two numbers between 1 and 99 that are used to represent sentiment, with a multiplier between 0 and 100 on a second one of the two sentiment numbers such that values between 1 and 9999 are possible. ### Claim 12 A computer-implemented method, comprising: by a computing system, generating relational sequences in trade space by performing principal component analysis (PCA) on raw data resulting in a UID comprising numbers derived from a dimensional reduction technique that provides relational meaning between the UID and other UIDs, wherein the trade space represents relative relationships in the raw data; mapping a solution space based on the generated relational sequences, by the computing system; replaying a plurality of scenarios, by the computing system, to determine pinch points where path structures are proximate to one another and more favorable outcomes can be jumped to; and generating a new solution path, by the computing system, based on one or more jumps to more favorable outcomes. ### Claim 13 (depends on 12) The computer-implemented method of claim 12, further comprising: seeding the plurality of scenarios with initial conditions, by the computing system, wherein the initial conditions are seeded with real world data; and initializing random states, by the computing system, to explore more preferable solutions. ### Claim 14 (depends on 12) The computer-implemented method of claim 12, further comprising: saving each relational sequence into a positive, negative, or neutral bin based on its outcome, by the computing system, wherein the neutral bin is a default for observations of natural phenomena in order to simulate natural phenomena that cannot be changed. ### Claim 15 (depends on 12) The computer-implemented method of claim 12, wherein a frequency with which a path including a given node is chosen is correlated to a number of times that the node has appeared during past simulations. ### Claim 16 (depends on 12) The computer-implemented method of claim 12, wherein when a negative node is inevitable on all paths, a chance that a jump is attempted is increased. ### Claim 17 A computer program embodied on a non-transitory computer-readable medium, the computer program configured to cause at least one processor to: generate relational sequences in trade space by performing principal component analysis (PCA) on raw data, wherein the trade space represents relative relationships in the raw data; map a solution space based on the generated relational sequences; replay a plurality of scenarios to determine pinch points where path structures are proximate to one another and more favorable outcomes can be jumped to; and generate a new solution path, by the computing system, based on one or more jumps to more favorable outcomes, wherein the PCA results in a UID comprising six numbers, with three of the six numbers resulting from principal components of a 2D wavelet transform of a respective image and another three of the six numbers resulting from principal components of a discrete cosine transform (DCT) of the respective image, which provides relational meaning between the UID and other UIDs. ### Claim 18 (depends on 17) The computer program of claim 17, further configured to: seed the plurality of scenarios with initial conditions, wherein the initial conditions are seeded with real world data; and initialize random states to explore more preferable solutions. ### Claim 19 (depends on 17) The computer program of claim 17, further configured to: save each relational sequence into a positive, negative, or neutral bin based on its outcome. ### Claim 20 (depends on 19) The computer program of claim 19, wherein the neutral bin is a default for observations of natural phenomena in order to simulate natural phenomena that cannot be changed.