19 results about "Problem solver" patented technology

The invention provides a software demand defect detection method based on a large language model and a satisfiability problem solver, and the method comprises the steps: constructing a domain knowledge base, guiding the large language model to extract key elements in a demand text through combining with a cue word project, and recognizing imperfect and ambiguous defects; and meanwhile, the demand is converted into a logic expression, and logic inconsistency is detected by using an SAT solver. According to the method, incremental updating of the knowledge base is supported, efficient retrieval and comparison are achieved through vectorization and Hash coding technologies, experiments show that the method is remarkably superior to a traditional technology in demand detection in the fields of finance and the like, defects can be accurately positioned, correction suggestions can be provided, and the efficiency and accuracy of demand analysis are effectively improved.

The invention provides a large language model vertical field multi-task objective question evaluation set automatic generation method, and relates to the technical field of artificial intelligence, and the method comprises the steps: for each text in a vertical field plain text data set, according to a plurality of preset task types, calling a problem solver to carry out task suitability discrimination on the text, and obtaining a task suitability discrimination result; when the judgment result is appropriate, generating at least one initial objective question stem of the task type; performing ambiguity detection on the initial objective question stem, and calling a question solver to rewrite the ambiguous initial objective question stem to obtain a rewritten objective question stem; and finally, calling a question solver to generate corresponding objective question options, objective question answers and objective question answer analyses for each rewritten objective question stem, and taking the objective question options, objective question answers and objective question answer analyses as an objective question evaluation set of a corresponding task type. By means of the method and device, the problems that in the prior art, time and labor are consumed in the large model evaluation dataset construction process, and automatic generation of the evaluation dataset in the specific vertical field cannot be achieved are solved.

The invention belongs to an intelligent highway pavement structure based on a self-sensing asphalt material and a monitoring system of the intelligent highway pavement structure, and discloses the intelligent highway pavement structure based on the self-sensing asphalt material and the monitoring system of the intelligent highway pavement structure. A pavement structural surface layer is made of a self-sensing asphalt mixture containing an amorphous silicon-based composite functional phase, the functional phase is prepared through a specific process, a nano bowl-shaped array is arranged on the surface of the functional phase, tungsten carbide nano crystal nucleuses are coated on the surface of the functional phase, and the functional phase and asphaltene form a non-equilibrium electron cloud overlapping system. The monitoring system is based on a non-contact microwave resonant cavity coupling principle, analyzes reflected signals, extracts characteristic vectors, and inverts real-time stress, strain, damage factors and temperature state parameters in combination with an inverse problem solver. A traditional sensor pre-embedding mode is abandoned, the pavement body and the sensing function are fused, the problems of implantation damage, life mismatching, low spatial resolution and the like are solved, and a new normal form is provided for intelligent road safety early warning and full-life management.

A spiking neural network (SNN) system is disclosed having spiking neurons associated with clause values associated with a satisfiability (SAT) problem. Each spiking neuron generates a voltage spike when on input voltage applied to each spiking neuron is increased above a spiking voltage threshold. The input voltage that is increased above the spiking voltage threshold is indicative that the corresponding clause value is satisfied. A controller applies the clause grid that includes clause values to a literal grid that includes literal values. The controller generates the input voltage that is applied to each spiking neuron based on each corresponding clause value associated with each spiking neuron that is applied to the literal values. The controller determines that each clause value is satisfied when the corresponding spiking neuron generates the voltage spike and that each clause value is unsatisfied when the corresponding spiking neuron fails to generate the voltage spike.

The invention provides a test method and system for a mathematical character question solver, and the system comprises a data preprocessing module which is used for integrating a multi-source mathematical character question data set, carrying out the standardized processing and marking of questions in the mathematical character question data set, and outputting a data instance with a unified data structure; the solver module is used for decomposing a solving algorithm into different basic function components as algorithm components, calling related algorithm components to form a corresponding solver according to a logic sequence of a to-be-tested solving algorithm, and solving a mathematical problem by using the solver based on an input data instance to generate a solving answer; and the perspective confusion comparison module is used for receiving the solving answer and comprehensively evaluating a solver corresponding to the algorithm to be measured through three dimensions of question type classification, key question solving steps and an algorithm component. According to the invention, multi-source data sets can be integrated, automatic and reproducible test workflows are provided, and fine-grained and multi-dimensional performance tests are supported.

The invention discloses a privacy protection elastic distributed solver design method and system for resisting mixed attacks in resource allocation and a storage medium, and the method comprises the steps: designing a solver in combination with steganography based on a classical distributed optimization framework for a mixed attack scene in which Byzantine attacks and eavesdropping attacks exist at the same time, and designing a privacy protection elastic distributed solver in combination with steganography; the privacy protection effect can be achieved while the Byzantine attack is resisted. Compared with a traditional resource allocation problem solver, the method has the capacity of processing hybrid attacks, and convergence errors of the algorithm are reduced. According to the method, two types of attacks can be confronted at the same time, malicious node interference is effectively inhibited through robust screening and an elastic updating mechanism, so that the system still keeps stable convergence with relatively low error in a mixed attack environment, privacy embedding with low calculation overhead is realized by introducing steganography and Hadamard transformation, and the method does not depend on heavy encryption or an MPC protocol, so that the method is simple and convenient to operate. And the real-time requirement of an intelligent power grid is met.

A method of resource management for transferring resources from providers to consumers is described. The method comprises the following steps. Supply parameters for the providers and demand parameters for the consumers are established, and constraints on the allocation of the resources are also established, as is an optimisation function for determining matches between providers and consumers. The optimisation function is solved for the established constraints using a constraint-based problem solver to determine matches between providers and consumers. Resources are then transferred from providers to consumers according to the determined matches. A computing system adapted to perform this method is also described, along with a power distribution system including such a computing system.

The invention relates to an artificial intelligence teaching system for guided learning and an application thereof, the artificial intelligence teaching system can provide an integrated intelligent collaborative learning platform taking cases as a center, and the system can create a complete clinical situation for learners, so that the learners can learn more clearly. A direct relation between morphological characteristics and disease diagnosis is helped to be established, so that the clinical comprehensive thinking ability is cultivated; and meanwhile, the learner is changed from a passive observer to an active and cooperative problem solver. Moreover, in order to solve the problem of lack of real-time professional guidance in collaborative learning, the system innovatively introduces an AI intelligent guidance module, so that interactive behaviors of learning groups are analyzed in real time, prompts, problems and feedbacks highly related to situations are provided, students are actively guided to focus on key features, potential error cognition is corrected, and learning efficiency is improved. And an immersive and efficient school effect of simulating expert guidance aside is achieved.

An object of the present invention is to provide a novel technology that accurately supports problem providers in solving their problems and further improves the benefits that problem providers can obtain. [Solution] In order to solve the above problems, the present invention provides a problem-solving support system that supports problem-solving by problem providers, the problem-solving support system comprising an index generation unit and a calculation unit, wherein the index generation unit generates a problem provider profit index that structurally organizes items for increasing the profit of the problem provider based on the problem provided by the problem provider, and the calculation unit calculates the problem provider profit that the problem provider will gain by providing the solution based on the problem provider profit index and the problem solver's solution information.

The present application provides a processing method and system for a mathematical problem solver based on deep learning. Since the problem-solving models in the geometry solver and the algebra solver are obtained by training using neural networks, one of the geometry solver and the algebra solver can be selected as a target solver according to the search question input by the user, and the search question can be input into the problem-solving model of the target solver for problem-solving processing, and the problem-solving result can be output to the user for viewing. This method does not require the storage of a question bank, and can perform intelligent solutions for various geometry and algebra problems, and the solution rate and accuracy are effectively improved.

The present invention provides a communication network topology planning method and system, comprising: modeling network planning problem instance parameters as a network planning graph signal X; and establishing a conversion relationship between a network topology solution and a graph adjacency matrix; inputting the obtained network planning signal X into a communication network planning problem solver based on a graph diffusion generation model, and combining the conversion relationship to obtain an optimal solution C of the network planning problem. ★ The present invention is a novel generative sampling solution that can obtain multiple high-quality network topology feasible solutions at once with low computational complexity, effectively improving the efficiency of solving network planning problems.

A computer-implemented method of interpreting a solution 12 of declarative problem codes 10, 11 is presented. The solution 12 and the natural language problem 20 with respect to the latter are provided to a machine learning system 2 that determines an auxiliary constraint 21 opposite to the premise of the problem 20. A declarative problem solver 1 identifies conflicts between these auxiliary constraints 21 and constraints of declarative problem codes 10, 11. The machine learning system 2 then determines an interpretation 31 based on conflict constraints of the declarative problem codes 10, 11.

A spiking neural network (SNN) system is disclosed having spiking neurons associated with clause values associated with a satisfiability (SAT) problem. Each spiking neuron generates a voltage spike when on input voltage applied to each spiking neuron is increased above a spiking voltage threshold. The input voltage that is increased above the spiking voltage threshold is indicative that the corresponding clause value is satisfied. A controller applies the clause grid that includes clause values to a literal grid that includes literal values. The controller generates the input voltage that is applied to each spiking neuron based on each corresponding clause value associated with each spiking neuron that is applied to the literal values. The controller determines that each clause value is satisfied when the corresponding spiking neuron generates the voltage spike and that each clause value is unsatisfied when the corresponding spiking neuron fails to generate the voltage spike.

A computer-implemented procedure for explaining a solution 12 to a declarative problem coding 10, 11 is presented. The solution 12 and a natural language question 20 related to it are provided to a machine learning system 2, which determines additional constraints 21 that contradict the premise of the question 20. The declarative problem solver 1 identifies conflicts between these auxiliary constraints 21 and the constraints of the declarative problem coding 10, 11. The machine learning system 2 then determines an explanation 31 based on the conflicting constraints of the declarative problem coding 10, 11.

A QUBO solving method for optimizing a function f(x) defined by a matrix Q and a vector p such 5 that and comprising the steps of: encrypting the function f(x) to be optimized by using an encryption matrix P and / or an encryption vector k; sending the encrypted version of the function to be optimized f′(x′) to a QUBO solver; receiving an encrypted version of the solution to the QUBO problem x′*, obtained by optimization of the encrypted version of the function f′(x′) performed by the QUBO solver; and decrypting the solution to the QUBO problem, thus obtaining x*.

This application provides a test question generation method, apparatus, electronic device, and storage medium. The method includes: parsing test question data to generate a conditional reasoning dependency graph representing the problem-solving logic chain, and extracting topological semantic fingerprints to construct a fingerprint database; generating a test question blueprint that meets the discrimination conditions through adversarial game between the generator and a discriminator simulating the cognitive state of the problem solver; calculating the cognitive entropy of the test question blueprint, and generating the target test question after completing threshold verification. This application can improve the quality of generated test questions.

The present invention discloses a non-autoregressive math problem solver based on a multi-way tree structure. On the basis of the existing problem encoding module and the target-driven multi-way tree generation module, a non-autoregressive target decomposition module is added to handle the unordered multi-branch decomposition work. First, the parent target \(E_g\) is fused with \(I\) positional encodings \(p_i\), \(i = 1, 2, \ldots, I\), then processed using the multi-head self-attention mechanism, and then a multi-head cross-attention module is used to connect the output of the multi-head self-attention module and the candidate set \(E_c\) to obtain the context representation. Finally, according to the context representation and the candidate set \(E_c\), the candidate characters in the most relevant candidate set \(E_c\) are selected through a pointer network to obtain the probability distribution matrix Ptr, which is used to calculate the cross-entropy loss during the training process and the predicted math problem expression during the prediction process. The present invention uses a non-autoregressive target decomposition module to handle the unordered multi-branch decomposition work, explores and captures the relationships between numbers, and improves the performance of the math problem solver.

A computer-implemented method of explaining a solution of a declarative problem encoding. The solution and a natural language question about the latter are provided to a machine learning system, which determines auxiliary constraints 21 opposing the questions premise. The declarative problem solver identifies conflicts between these auxiliary constraints and the constraints of the declarative problem encoding. The machine learning system then determines an explanation based on the conflicting constraints of the declarative problem encoding.

The present application relates to the technical field of chip, especially relates to a random constraint problem solver core determination method, electronic equipment and medium, through constructing a solver core determination model with input of network attribute vector, node attribute vector and adjacent network matrix and output of each solver core weight, inputting the input data in the form of network attribute vector, node attribute vector and adjacent network matrix corresponding to the to-be-solved random constraint problem into the solver core determination model, the solver core corresponding to the to-be-solved random constraint problem can be determined, so that all inputted random constraint problems can be matched to the optimal solver core. The present application improves the solving efficiency and accuracy of random constraint problems.