Multi-objectivization of reinforcement learning problems by reward shaping

T Brys, A Harutyunyan, P Vrancx… - … joint conference on …, 2014 - ieeexplore.ieee.org
Multi-objectivization is the process of transforming a single objective problem into a multi-
objective problem. Research in evolutionary optimization has demonstrated that the addition …

MMO: Meta multi-objectivization for software configuration tuning

P Chen, T Chen, M Li - IEEE Transactions on Software …, 2024 - ieeexplore.ieee.org
Software configuration tuning is essential for optimizing a given performance objective (eg,
minimizing latency). Yet, due to the software's intrinsically complex configuration landscape …

A Switch-and-Restart Algorithm with Exponential Restart Strategy for Objective Selection and its Runtime Analysis

M Buzdalov - 2014 13th International Conference on Machine …, 2014 - ieeexplore.ieee.org
There exist optimization problems with the target objective, which is to be optimized, and
several extra objectives, which may or may not be helpful in the optimization process. This …

Power electronics' cool new flavor [news]

R Stevenson - IEEE Spectrum, 2016 - ieeexplore.ieee.org
Ideally, the electronic components that route electricity through power supplies, inverters,
and electric motors are cheap, efficient, and capable of handling high voltages. Judged in …

Can OneMax help optimizing LeadingOnes using the EA+ RL method?

M Buzdalov, A Buzdalova - 2015 IEEE Congress on …, 2015 - ieeexplore.ieee.org
There exist optimization problems with the target objective, which is to be optimized, and
several extra objectives, which can be helpful in the optimization process. The EA+ RL …

Worst-case execution time test generation for augmenting path maximum flow algorithms using genetic algorithms

V Arkhipov, M Buzdalov… - 2013 12th International …, 2013 - ieeexplore.ieee.org
Worst-case execution time tests can be tricky to create for various computer science
algorithms. To reduce the amount of human effort, authors suggest using search-based …

Hard test generation for augmenting path maximum flow algorithms using genetic algorithms: Revisited

M Buzdalov, A Shalyto - 2015 IEEE Congress on Evolutionary …, 2015 - ieeexplore.ieee.org
To estimate performance of computer science algorithms reliably, one has to create worst-
case execution time tests. For certain algorithms this task can be difficult. To reduce the …

Risk-sensitivity through multi-objective reinforcement learning

K Van Moffaert, T Brys, A Nowé - 2015 IEEE Congress on …, 2015 - ieeexplore.ieee.org
Usually in reinforcement learning, the goal of the agent is to maximize the expected return.
However, in practical applications, algorithms that solely focus on maximizing the mean …

Модели и методы процесса автоматизации многофункциональных виртуальных тренажеров для подготовки специалистов рыбопромыслового флота

НП Сметюх - 2019 - elibrary.ru
Использование виртуальных тренажеров является важнейшим трендом развития
современного образования, в том числе для непрерывного образования специалистов …

Vulture voyeur a sensor-packed egg monitors nests from the inside

B Feng, B Liu, K Pan - IEEE Spectrum, 2016 - ieeexplore.ieee.org
Vultures are nature's garbage collectors, helping the environment by consuming dead
animal carcasses. In this way, they are essential in stopping the spread of diseases such as …