Abstract: An example system includes a processor to receive a data set for training a machine learning model. The processor can train the machine learning model on the data set. The processor can also translate the machine learning model into constraint satisfaction problem (CSP) variables and constraints. The processor can generate fabricated data based on the CSP variables and constraints.

Abstract: A computer-implemented method comprising, automatically: analyzing a machine learning dataset which comprises multiple datapoints, to deduce constraints on features of the datapoints; generating a first set of CSP (Constraint Satisfaction Problem) rules expressing the constraints; based on a machine learning model which was trained on the dataset, generating a second set of CSP rules that define one or more perturbation candidates among the features of one of the datapoints; formulating a CSP based on the first and second sets of CSP rules; solving the formulated CSP using a solver; and using the solution of the CSP as a counterfactual explanation of a prediction made by the machine learning model with respect to the one datapoint.

Abstract: Technical solutions are described for reusing a solution for a test. An example method includes building, by a processor, a solution cache including a tree structure representative of a plurality of solutions, which stores a key configurable immediate value of a previous solution as a node, the previous solution as a leaf node of the tree, and an edge from the node indicative of a value of the key configurable immediate value at the node. The method includes traversing nodes of the tree structure in the solution cache to identify key configurable immediate values of a previous solution identical to configurable immediate values from the test by identifying edges associated values identical to those from the test. In response to reaching a leaf node of the tree structure, using the solution(s) at the leaf node as a solution of the test.

Abstract: A method, apparatus and product for solving CSP comprising vectors of unknown size. The method comprises generating a structural skeleton tree of a problem description, wherein the structural skeleton tree comprises a node representing a vector of unknown size and a node representing a size of the vector; determining a vector size Constraint Satisfaction Problem (CSP) based on the structural skeleton tree, wherein said determining comprises projecting over-approximated constraints on the size of the vector based on operators used on the vector or elements thereof; solving the vector size CSP to determine the size of the vector; modifying the structural skeleton tree to set the size of the vector and to include nodes for each element in the vector, whereby obtaining a CSP; and solving the CSP.

Abstract: A method and product for solving a Constraint Satisfaction Problem (CSP) having a constraint associated with a string variable, comprising: defining a string variable using a string domain data structure representing a domain of string values for string variables, the structure represents the domain as a Deterministic Finite Automaton (DFA) with no back loops longer than 1 and edges representing characters; defining a constraint for the CSP, wherein the constraint involves the variable and is to be complied with by a solution to the CSP; invoking a CSP solver using the string variable while complying with the constraint, and to invoke operations performed over the domain, wherein the solver: propagates values into the domain of string values, wherein the value propagation reduces the domain size; or selects values from the domain thus reducing the domain to a singleton domain, whereby value propagation to domains of other variables is invoked by the solver.

Abstract: A method and product for solving a Constraint Satisfaction Problem (CSP) having a constraint associated with a string variable, comprising: defining a string variable using a string domain data structure representing a domain of string values for string variables, the structure represents the domain as a Deterministic Finite Automaton (DFA) with no back loops longer than 1 and edges representing characters; defining a constraint for the CSP, wherein the constraint involves the variable and is to be complied with by a solution to the CSP; invoking a CSP solver using the string variable while complying with the constraint, and to invoke operations performed over the domain, wherein the solver: propagates values into the domain of string values, wherein the value propagation reduces the domain size; or selects values from the domain thus reducing the domain to a singleton domain, whereby value propagation to domains of other variables is invoked by the solver.

Abstract: Technical solutions are described for reusing a solution for a test. An example method includes building, by a processor, a solution cache including a tree structure representative of a plurality of solutions, which stores a key configurable immediate value of a previous solution as a node, the previous solution as a leaf node of the tree, and an edge from the node indicative of a value of the key configurable immediate value at the node. The method includes traversing nodes of the tree structure in the solution cache to identify key configurable immediate values of a previous solution identical to configurable immediate values from the test by identifying edges associated values identical to those from the test. In response to reaching a leaf node of the tree structure, using the solution(s) at the leaf node as a solution of the test.

Abstract: An apparatus and a computer-implemented method performed by a computerized device, comprising: generating a collection of test data for testing one or more domains, wherein the test data is useful for post-silicon verification of hardware devices; selecting a subset of the collection of test data in accordance with a hardware device to be tested and at least one of the domains to be tested with respect to the hardware device; and indexing the subset of the collection of test data to obtain an indexed collection.

Abstract: A computer-implemented method, apparatus and product for solving a Constraint Satisfaction Problem (CSP). The computer-implemented method, performed by a computerized device, comprising: obtaining the CSP comprising a set of variables, a set of corresponding domains, and a set of constraints; and solving the CSP by iteratively: limiting the domains to exclude values that cannot satisfy one or more constraints of the set of constraints; heuristically selecting a variable out of the set of variables, wherein said heuristically selecting is based on constraint semantics of a constraint of the set of constraints; and limiting a corresponding domain of the selected variable.

Abstract: A computer-implemented method, apparatus and product for solving a Constraint Satisfaction Problem (CSP). The computer-implemented method, performed by a computerized device, comprising: obtaining the CSP comprising a set of variables, a set of corresponding domains, and a set of constraints; and solving the CSP by iteratively: limiting the domains to exclude values that cannot satisfy one or more constraints of the set of constraints; heuristically selecting a variable out of the set of variables, wherein said heuristically selecting is based on constraint semantics of a constraint of the set of constraints; and limiting a corresponding domain of the selected variable.

Abstract: A CSP is modified to include an auxiliary variable. A component of a constraint of the CSP is modified to refer to the auxiliary variable instead of referring to a pair of variables. A domain for the auxiliary variable may be defined based on domains of the pair of variables. A binding constraint is added to the CSP to bind the value of the auxiliary variable with the values of the pair of variables. The modified CSP is solved using a CSP solver.

Abstract: An apparatus and a computer-implemented method performed by a computerized device, comprising: generating a collection of test data for testing one or more domains, wherein the test data is useful for post-silicon verification of hardware devices; selecting a subset of the collection of test data in accordance with a hardware device to be tested and at least one of the domains to be tested with respect to the hardware device; and indexing the subset of the collection of test data to obtain an indexed collection.

Abstract: A CSP is modified to include an auxiliary variable. A component of a constraint of the CSP is modified to refer to the auxiliary variable instead of referring to a pair of variables. A domain for the auxiliary variable may be defined based on domains of the pair of variables. A binding constraint is added to the CSP to bind the value of the auxiliary variable with the values of the pair of variables. The modified CSP is solved using a CSP solver.

Abstract: Apparatus for automatically generating test programs is provided. The apparatus includes a test generator, which is adapted to receive a description of a system under test, expressed in terms of variables associated with the system and conditional constraints including semantics applied to the variables, to receive a definition of an event to be tested in the system, to generate an ECondCSP over the variables responsively to the definition of the event and to the conditional constraints, such that at least some of the semantics of the conditional constraints are preserved in the ECondCSP when one or more of the variables to which the semantics are applied are inactive, and to solve the ECondCSP to generate a test case for the system.

Abstract: A method for automatically generating test programs includes receiving a description of a system under test, expressed in terms of variables associated with the system and conditional constraints including semantics applied to the variables, and receiving a definition of an event to be tested in the system. The method generates an ECondCSP over the variables responsively to the definition of the event and to the conditional constraints, such that at least some of the semantics of the conditional constraints are preserved in the ECondCSP when one or more of the variables to which the semantics are applied are inactive. The ECondCSP is solved to generate a test case for the system.

Abstract: A method for design verification includes running a simulation of a design in a simulation environment, which comprises a stimuli generator for providing inputs to the design during the simulation. Respective measures of quality are computed for at least some of the simulation states in a sequence of states generated by the environment. State data are saved with respect to at least one of the simulation states. The state data include indications both of the respective simulated state and of the respective environment state. Responsively to the respective measures of quality, the saved state data are recalled so as to restart the simulation from the at least one of the simulation states by returning the design to the respective simulated state and returning the simulation environment to the respective environment state.

Abstract: A method for automatically generating test programs includes receiving a description of a system under test, expressed in terms of variables associated with the system and conditional constraints including semantics applied to the variables, and receiving a definition of an event to be tested in the system. The method generates an ECondCSP over the variables responsively to the definition of the event and to the conditional constraints, such that at least some of the semantics of the conditional constraints are preserved in the ECondCSP when one or more of the variables to which the semantics are applied are inactive. The ECondCSP is solved to generate a test case for the system.