Aspects of the present disclosure relate to confidence volumes for earth modeling using machine learning. A method includes receiving detected data, wherein the detected data includes formation attributes relating to one or more depth points along or near a wellbore. The method further includes providing inputs to a plurality of machine learning models based on the detected data. The method further includes receiving output values from the plurality of machine learning models based on the inputs. The method further includes determining a measure of variance among the output values. The method further includes generating a confidence indicator related to the output values based on the measure of variance.

A learning processing unit (24) causes a second neural network (NN) (18) to be trained, with a hyperparameter set of a first NN (16) accepted as input, so as to output post-learning performance that is the performance of a trained first NN (16) to which the hyperparameter set is set. A GA processing unit (26) adjusts the hyperparameter set of the first NN (16) by a genetic algorithm, with the hyperparameter set of the first NN (16) handled as entity, the fitness of said algorithm being configured to be a value that corresponds to the post-learning performance of the first NN (16) to which the hyperparameter set is set. In processing in each generation of the genetic algorithm, the post-learning performance of the first NN (16) corresponding to each hyperparameter is acquired using the second NN (18).

A learning processing unit (24) causes a second neural network (NN) (18) to be trained, with a hyperparameter set of a first NN (16) accepted as input, so as to output post-learning performance that is the performance of a trained first NN (16) to which the hyperparameter set is set. A GA processing unit (26) adjusts the hyperparameter set of the first NN (16) by a genetic algorithm, with the hyperparameter set of the first NN (16) handled as entity, the fitness of said algorithm being configured to be a value that corresponds to the post-learning performance of the first NN (16) to which the hyperparameter set is set. In processing in each generation of the genetic algorithm, the post-learning performance of the first NN (16) corresponding to each hyperparameter is acquired using the second NN (18).

The present disclosure describes a system and method for improving the way computing devices execute genetic algorithms. A fuzzy logic controller takes various properties of the genetic algorithm (such as the diversity of the population, the performance history of the algorithm in terms of time-efficiency and/or effectiveness at improving the best fitness function results, and available computing resources) to dynamically manage the parameters of the genetic algorithm. In some embodiments, the fuzzy inference system that provides parameters to the genetic algorithm is itself controlled by another fuzzy inference system.

A method includes: generating a plurality of individuals of a current generation in accordance with a plurality of individuals of a previous generation to acquire values of an objective function for individuals each representing a variable by evolutionary computation; calculating, for each of partial individuals of the plurality of individuals of the current generation generated by the generating processing, a first value of the objective function by a predetermined method; approximately calculating, for each of the plurality of individuals of the current generation, a second value of the objective function with lower precision than the predetermined method; computing a fitness difference representing a difference between the plurality of individuals of the current generation in accordance with the first value or the second value; and controlling the precision of the approximate calculation based on the fitness difference and a precision difference between the first value and the second value.

An electronic device such as an accessory-mountable robot is provided. The electronic device changes functional properties thereof in accordance with a mounted accessory. In an embodiment, the electronic device detects mounting of at least one accessory and identifies accessory characteristics associated with the at least one accessory. Then, the electronic device determines properties of the electronic device associated with the at least one accessory, based on the accessory characteristics, and changes the properties of the electronic device, based on the determined properties. Also, the electronic device outputs at least one of a visual element, an auditory element, or a tactile element associated with the at least one accessory, based on the changed properties.

Methods, non-transitory machine readable media, and network stack analysis devices that generate optimized preferred provider organization (PPO) network stacks are disclosed. With this technology, electronic transactions are applied to each of a first plurality of network stacks to determine a cost reduction value for each of the first network stacks. Each of the first network stacks includes an ordered subset of networks. The first network stacks are resampled based on the determined cost reduction values. A determination is made when one or more convergence criteria are met by the resampled first network stacks. When the determination indicates that the convergence criteria are not met by the resampled first network stacks, one or more of the first network stacks are modified based on genetic crossover or mutation operation(s) to generate a second plurality of network stacks. The application, resampling, and determination are then repeated for the second network stacks.

Techniques regarding determining a three-dimensional structure of a heteropolymer are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a polymer folding component that can generate a course-grained model to determine a three-dimensional structure of a heteropolymer based on a first qubit registry that encodes a conformation of the heteropolymer on a lattice and a second qubit registry that encodes an interaction distance between monomers comprised within the heteropolymer.

The subject technology is related to methods and apparatus for training a set of regression machine learning models with a training set to produce a set of predictive values for a pending manufacturing request, the training set including data extracted from a set of manufacturing transactions submitted by a set of entities of a supply chain. A multi-objective optimization model is implemented to (1) receive an input including the set of predictive values and a set of features of a physical object, and (2) generate an output with a set of attributes associated with a manufacture of the physical object in response to receiving the input, the output complying with a multi-objective condition satisfied in the multi-objective optimization model.

The subject technology is related to methods and apparatus for training a set of regression machine learning models with a training set to produce a set of predictive values for a pending manufacturing request, the training set including data extracted from a set of manufacturing transactions submitted by a set of entities of a supply chain. A multi-objective optimization model is implemented to (1) receive an input including the set of predictive values and a set of features of a physical object, and (2) generate an output with a set of attributes associated with a manufacture of the physical object in response to receiving the input, the output complying with a multi-objective condition satisfied in the multi-objective optimization model.