This past Monday I released v5.2.0 of Chips-n-Salsa (link to project below). That release added implementation of (1+1)-EA (evolutionary algorithm). I'm currently working on integrating and/or reimplementing my code for various crossover operators for evolving permutations into the library. Much of my existing code for this is from prior research projects with code written for very specific scheduling and other optimization problems. So in some cases at least partially reimplementing to be problem-independent as I'm integrating with the Chips-n-Salsa library. In some cases, my prior code I'm digging into I originally wrote as long ago as 20 years ago.
The Chips-n-Salsa library includes implementations of several stochastic local search algorithms, including simulated annealing, hill climbers, as well as constructive search algorithms such as stochastic sampling; and now also includes genetic algorithms as well as evolutionary algorithms more generally. It includes several classes for representing solutions to a variety of optimization problems. For example, the library includes a BitVector class that implements vectors of bits, as well as classes for representing solutions to problems where we are searching for an optimal vector of integers or reals. For each of the built-in representations, the library provides the most common mutation operators and crossover operators for use with evolutionary algorithms. The library provides extensive support for permutation optimization problems, including implementations of many different mutation operators for permutations, and utilizing the efficiently implemented Permutation class of the JavaPermutationTools (JPT) library. Chips-n-Salsa is customizable, making extensive use of generic types, enabling using the library to optimize other types of representations beyond what is provided in the library. It is hybridizable, providing support for integrating multiple forms of local search (e.g., using a hill climber on a solution generated by simulated annealing), creating hybrid mutation operators (e.g., local search using multiple mutation operators), and classes that support running more than one type of search for the same problem concurrently using multiple threads as a form of algorithm portfolio. Chips-n-Salsa is iterative, with support for multistart metaheuristics, including implementations of several restart schedules for varying the run lengths across the restarts. It also supports parallel execution of multiple instances of the same, or different, stochastic local search algorithms for an instance of a problem to accelerate the search process. The library supports self-adaptive search in a variety of ways, such as including implementations of adaptive annealing schedules for simulated annealing, such as the Modified Lam schedule, implementations of the simpler annealing schedules but which self-tune the initial temperature and other parameters, and restart schedules that adapt to run length.
chips-n-salsa.cicirello.org
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
This past Monday I released v5.2.0 of Chips-n-Salsa (link to project below). That release added implementation of (1+1)-EA (evolutionary algorithm). I'm currently working on integrating and/or reimplementing my code for various crossover operators for evolving permutations into the library. Much of my existing code for this is from prior research projects with code written for very specific scheduling and other optimization problems. So in some cases at least partially reimplementing to be problem-independent as I'm integrating with the Chips-n-Salsa library. In some cases, my prior code I'm digging into I originally wrote as long ago as 20 years ago.
Chips-n-Salsa - A Java library of customizable, hybridizable, iterative, parallel, stochastic, and self-adaptive local search algorithms
The Chips-n-Salsa library includes implementations of several stochastic local search algorithms, including simulated annealing, hill climbers, as well as constructive search algorithms such as stochastic sampling; and now also includes genetic algorithms as well as evolutionary algorithms more generally. It includes several classes for representing solutions to a variety of optimization problems. For example, the library includes a BitVector class that implements vectors of bits, as well as classes for representing solutions to problems where we are searching for an optimal vector of integers or reals. For each of the built-in representations, the library provides the most common mutation operators and crossover operators for use with evolutionary algorithms. The library provides extensive support for permutation optimization problems, including implementations of many different mutation operators for permutations, and utilizing the efficiently implemented Permutation class of the JavaPermutationTools (JPT) library. Chips-n-Salsa is customizable, making extensive use of generic types, enabling using the library to optimize other types of representations beyond what is provided in the library. It is hybridizable, providing support for integrating multiple forms of local search (e.g., using a hill climber on a solution generated by simulated annealing), creating hybrid mutation operators (e.g., local search using multiple mutation operators), and classes that support running more than one type of search for the same problem concurrently using multiple threads as a form of algorithm portfolio. Chips-n-Salsa is iterative, with support for multistart metaheuristics, including implementations of several restart schedules for varying the run lengths across the restarts. It also supports parallel execution of multiple instances of the same, or different, stochastic local search algorithms for an instance of a problem to accelerate the search process. The library supports self-adaptive search in a variety of ways, such as including implementations of adaptive annealing schedules for simulated annealing, such as the Modified Lam schedule, implementations of the simpler annealing schedules but which self-tune the initial temperature and other parameters, and restart schedules that adapt to run length.