scholarly journals A Comparison of Global Search Algorithms for Continuous Black Box Optimization

2012 ◽  
Vol 20 (4) ◽  
pp. 509-541 ◽  
Author(s):  
Petr Pošík ◽  
Waltraud Huyer ◽  
László Pál
Keyword(s):  
State Of The Art ◽  
Search Space ◽  
Search Algorithms ◽  
Black Box ◽  
Search Method ◽  
The Other ◽  
Systematic Sampling ◽  
Careful Selection ◽  
Local Optimizer ◽  
Selection Of

Four methods for global numerical black box optimization with origins in the mathematical programming community are described and experimentally compared with the state of the art evolutionary method, BIPOP-CMA-ES. The methods chosen for the comparison exhibit various features that are potentially interesting for the evolutionary computation community: systematic sampling of the search space (DIRECT, MCS) possibly combined with a local search method (MCS), or a multi-start approach (NEWUOA, GLOBAL) possibly equipped with a careful selection of points to run a local optimizer from (GLOBAL). The recently proposed “comparing continuous optimizers” (COCO) methodology was adopted as the basis for the comparison. Based on the results, we draw suggestions about which algorithm should be used depending on the available budget of function evaluations, and we propose several possibilities for hybridizing evolutionary algorithms (EAs) with features of the other compared algorithms.

scholarly journals Random search with k -prototypes algorithm for clustering mixed datasets

2011 ◽  
Vol 467 (2132) ◽  
pp. 2387-2403 ◽  
Author(s):  
Duc-Truong Pham ◽  
Maria M. Suarez-Alvarez ◽  
Yuriy I. Prostov
Keyword(s):  
Optimization Algorithm ◽  
Random Search ◽  
Search Space ◽  
Population Based ◽  
The Other ◽  
Bees Algorithm ◽  
Neighbourhood Search ◽  
Promising Solution ◽  
Number Of Iterations ◽  
Selection Of

A new algorithm to cluster datasets with mixed numerical and categorical values is presented. The algorithm, called RANKPRO (random search with k -prototypes algorithm), combines the advantages of a recently introduced population-based optimization algorithm called the bees algorithm (BA) and k -prototypes algorithm. The BA works with elite and good solutions, and continues to look for other possible extrema solutions keeping the number of testing points constant. However, the improvement of promising solutions by the BA may be time-consuming because it is based on random neighbourhood search. On the other hand, an application of the k -prototypes algorithm to a promising solution may be very effective because it improves the solution at each iteration. The RANKPRO algorithm balances two objectives: it explores the search space effectively owing to random selection of new solutions, and improves promising solutions fast owing to employment of the k -prototypes algorithm. The efficiency of the new algorithm is demonstrated by clustering several datasets. It is shown that in the majority of the considered datasets when the average number of iterations that the k -prototypes algorithm needs to converge is over 10, the RANKPRO algorithm is more efficient than the k -prototypes algorithm.

Hardware and Software for Multimedia Development

2010 ◽  
pp. 60-68
Author(s):  
Manjit Singh Sidhu
Keyword(s):  
Design Process ◽  
Multimedia Learning ◽  
Cost Effective ◽  
The Other ◽  
Careful Selection ◽  
High Quality ◽  
Multimedia Development ◽  
The Subject ◽  
Audio Video ◽  
Selection Of

The aforementioned statement is essential to ponder when it comes to multimedia hardware and software consideration. Computers are now making it possible to blend or integrate together multimedia elements such as audio, video, graphics, images and animation into a single learning package. However, blending these multimedia elements together to develop a learning package does not mean that student’s proficiency in the subject matter could be enhanced. Furthermore it is not necessary to convert the entire textbook into a full working multimedia package for students to learn. Selected problems that are difficult to explain to the students from the textbook could be more appropriate and beneficial to the students if shown in the form of motion. On the other hand selecting appropriate multimedia elements and authoring tools could be difficult tasks for a new multimedia author. This is because proper development design process and careful selection of multimedia elements should be used appropriately to develop a high quality and cost effective package that could engage learners in their learning. Understanding the overall development process is an essential part for a multimedia learning package.

scholarly journals TextNAS: A Neural Architecture Search Space Tailored for Text Representation

2020 ◽  
Vol 34 (05) ◽  
pp. 9242-9249
Author(s):  
Yujing Wang ◽  
Yaming Yang ◽  
Yiren Chen ◽  
Jing Bai ◽  
Ce Zhang ◽  
...  
Keyword(s):  
Text Classification ◽  
Network Architecture ◽  
State Of The Art ◽  
Search Space ◽  
Search Algorithms ◽  
Text Representation ◽  
Neural Architecture ◽  
Good Potential ◽  
Automatic Search ◽  
Public Datasets

Learning text representation is crucial for text classification and other language related tasks. There are a diverse set of text representation networks in the literature, and how to find the optimal one is a non-trivial problem. Recently, the emerging Neural Architecture Search (NAS) techniques have demonstrated good potential to solve the problem. Nevertheless, most of the existing works of NAS focus on the search algorithms and pay little attention to the search space. In this paper, we argue that the search space is also an important human prior to the success of NAS in different applications. Thus, we propose a novel search space tailored for text representation. Through automatic search, the discovered network architecture outperforms state-of-the-art models on various public datasets on text classification and natural language inference tasks. Furthermore, some of the design principles found in the automatic network agree well with human intuition.

scholarly journals Online Bridged Pruning for Real-Time Search with Arbitrary Lookaheads

2017 ◽  
Author(s):  
Carlos Hernandez ◽  
Adi Botea ◽  
Jorge A. Baier ◽  
Vadim Bulitko
Keyword(s):  
Real Time ◽  
Recent Progress ◽  
State Of The Art ◽  
Search Algorithm ◽  
Search Space ◽  
Poor Quality ◽  
Search Algorithms ◽  
Arbitrary Size ◽  
Order Of Magnitude ◽  
And Robotics

Real-time search algorithms are relevant to time-sensitive decision-making domains such as video games and robotics. In such settings, the agent is required to decide on each action under a constant time bound, regardless of the search space size. Despite recent progress, poor-quality solutions can be produced mainly due to state re-visitation. Different techniques have been developed to reduce such a re-visitation with state pruning showing promise. In this paper, we propose a novel pruning approach applicable to the wide class of real-time search algorithms. Given a local search space of arbitrary size, our technique aggressively prunes away all states in its interior, possibly adding new edges to maintain the connectivity of the search space frontier. An experimental evaluation shows that our pruning often improves the performance of a base real-time search algorithm by over an order of magnitude. This allows our implemented system to outperform state-of-the-art real-time search algorithms used in the evaluation.

scholarly journals Online Relaxation Refinement for Satisficing Planning: On Partial Delete Relaxation, Complete Hill-Climbing, and Novelty Pruning

2022 ◽  
Vol 73 ◽  
Author(s):  
Maximilian Fickert ◽  
Jörg Hoffmann
Keyword(s):  
Local Minimum ◽  
State Of The Art ◽  
Convergence Property ◽  
Search Space ◽  
Search Algorithms ◽  
Hill Climbing ◽  
Ai Planning ◽  
Best First Search ◽  
Refinement Procedure ◽  
Hill Climbing Algorithms

In classical AI planning, heuristic functions typically base their estimates on a relaxation of the input task. Such relaxations can be more or less precise, and many heuristic functions have a refinement procedure that can be iteratively applied until the desired degree of precision is reached. Traditionally, such refinement is performed offline to instantiate the heuristic for the search. However, a natural idea is to perform such refinement online instead, in situations where the heuristic is not sufficiently accurate. We introduce several online-refinement search algorithms, based on hill-climbing and greedy best-first search. Our hill-climbing algorithms perform a bounded lookahead, proceeding to a state with lower heuristic value than the root state of the lookahead if such a state exists, or refining the heuristic otherwise to remove such a local minimum from the search space surface. These algorithms are complete if the refinement procedure satisfies a suitable convergence property. We transfer the idea of bounded lookaheads to greedy best-first search with a lightweight lookahead after each expansion, serving both as a method to boost search progress and to detect when the heuristic is inaccurate, identifying an opportunity for online refinement. We evaluate our algorithms with the partial delete relaxation heuristic hCFF, which can be refined by treating additional conjunctions of facts as atomic, and whose refinement operation satisfies the convergence property required for completeness. On both the IPC domains as well as on the recently published Autoscale benchmarks, our online-refinement search algorithms significantly beat state-of-the-art satisficing planners, and are competitive even with complex portfolios.

scholarly journals Approaches in Intelligent Music Production

Arts ◽  
2019 ◽  
Vol 8 (4) ◽  
pp. 125 ◽  
Author(s):  
David Moffat ◽  
Mark B. Sandler
Keyword(s):  
Intelligent Systems ◽  
State Of The Art ◽  
Autonomous Systems ◽  
Black Box ◽  
The Other ◽  
New Developments ◽  
Music Production ◽  
The Past ◽  
Digital Modelling ◽  
Key Aspects

Music production technology has made few advancements over the past few decades. State-of-the-art approaches are based on traditional studio paradigms with new developments primarily focusing on digital modelling of analog equipment. Intelligent music production (IMP) is the approach of introducing some level of artificial intelligence into the space of music production, which has the ability to change the field considerably. There are a multitude of methods that intelligent systems can employ to analyse, interact with, and modify audio. Some systems interact and collaborate with human mix engineers, while others are purely black box autonomous systems, which are uninterpretable and challenging to work with. This article outlines a number of key decisions that need to be considered while producing an intelligent music production system, and identifies some of the assumptions and constraints of each of the various approaches. One of the key aspects to consider in any IMP system is how an individual will interact with the system, and to what extent they can consistently use any IMP tools. The other key aspects are how the target or goal of the system is created and defined, and the manner in which the system directly interacts with audio. The potential for IMP systems to produce new and interesting approaches for analysing and manipulating audio, both for the intended application and creative misappropriation, is considerable.

Junction characteristics of energy wave methods

1997 ◽  
Vol 211 (1) ◽  
pp. 25-35 ◽  
Author(s):  
T Chen ◽  
R F Boucher
Keyword(s):  
Mixed Type ◽  
Constitutive Relations ◽  
Type Systems ◽  
The Other ◽  
Careful Selection ◽  
Physical Systems ◽  
Scattering Coefficients ◽  
Wave Methods ◽  
System Topology ◽  
Selection Of

Physical dynamical systems may be modelled by connecting wave-transmitting elements using junctions that depict the relations of energy transmission. Elements transmit wave energy from one end to the other end and junctions reallocate energy to those attached elements. The scattering coefficients of junctions may be derived by satisfying the continuity constitutive relations and energy conservations. The scattering coefficients expressed in general forms of 13 different junctions demonstrated in this paper permit systems with multiple energy domains to be modelled and are applicable to modelling distributed, lumped and mixed type systems. Careful selection of the correct types of junctions and construction of the system topology are keys to successfully modelling physical systems.

scholarly journals Bayes-Adaptive Monte-Carlo Planning and Learning for Goal-Oriented Dialogues

2020 ◽  
Vol 34 (05) ◽  
pp. 7994-8001
Author(s):  
Youngsoo Jang ◽  
Jongmin Lee ◽  
Kee-Eung Kim
Keyword(s):  
Reinforcement Learning ◽  
Task Performance ◽  
State Of The Art ◽  
Search Space ◽  
Difficult Problem ◽  
The Other ◽  
Adaptive Planning ◽  
Adaptive Monte Carlo ◽  
Planning Algorithm ◽  
Dialogue Planning

We consider a strategic dialogue task, where the ability to infer the other agent's goal is critical to the success of the conversational agent. While this problem can be naturally formulated as Bayesian planning, it is known to be a very difficult problem due to its enormous search space consisting of all possible utterances. In this paper, we introduce an efficient Bayes-adaptive planning algorithm for goal-oriented dialogues, which combines RNN-based dialogue generation and MCTS-based Bayesian planning in a novel way, leading to robust decision-making under the uncertainty of the other agent's goal. We then introduce reinforcement learning for the dialogue agent that uses MCTS as a strong policy improvement operator, casting reinforcement learning as iterative alternation of planning and supervised-learning of self-generated dialogues. In the experiments, we demonstrate that our Bayes-adaptive dialogue planning agent significantly outperforms the state-of-the-art in a negotiation dialogue domain. We also show that reinforcement learning via MCTS further improves end-task performance without diverging from human language.

scholarly journals NACRE - A Nogood And Clause Reasoning Engine

10.29007/dxnb ◽  
2020 ◽  
Author(s):  
Gael Glorian ◽  
Jean-Marie Lagniez ◽  
Christophe Lecoutre
Keyword(s):  
Learning Strategies ◽  
Data Structures ◽  
State Of The Art ◽  
Search Algorithms ◽  
Black Box ◽  
General Description ◽  
Modular Architecture ◽  
Search Methods ◽  
Constraint Solver ◽  
Reasoning Engine

NACRE, for Nogood And Clause Reasoning Engine, is a constraint solver written in C++. It is based on a modular architecture designed to work with generic constraints while implementing several state-of-the-art search methods and heuristics. Interestingly, its data structures have been carefully designed to play around nogoods and clauses, making it suit- able for implementing learning strategies. NACRE was submitted to the CSP MiniTrack of the 2018 and 2019 XCSP3 [8] competitions where it took the first place. This paper gives a general description of NACRE as a framework. We present its kernel, the available search algorithms, and the default settings (notably, used for XCSP3 competitions), which makes NACRE efficient in practice when used as a black-box solver.

scholarly journals PlotsOfData – a web app for visualizing data together with its summaries

2018 ◽  
Author(s):  
Marten Postma ◽  
Joachim Goedhart
Keyword(s):  
Data Structure ◽  
State Of The Art ◽  
The Other ◽  
Raw Data ◽  
Independent Evaluation ◽  
R Shiny ◽  
Web App ◽  
Data Visualizations ◽  
Data Summaries ◽  
Selection Of

AbstractReporting of the actual data in graphs and plots increases transparency and enables independent evaluation. On the other hand, data summaries are often used in graphs since they aid interpretation. State-of-the art data visualizations can be made with the ggplot2 package, which uses the ideas of a ‘grammar of graphics’ to generate a graphic from multiple layers of data. However, ggplot2 requires coding skills and an understanding of the tidy data structure. To democratize state-of-the-art data visualization of raw data with a selection of statistical summaries, a web app was written using R/shiny that uses the ggplot2 package for generating plots. A multilayered approach together with adjustable transparency offers a unique flexibility, enabling users can to choose how to display the data and which of the data summaries to add. Four data summaries are provided, mean, median, boxplot, violinplot, to accommodate several types of data distributions. In addition, 95% confidence intervals can be added for visual inferences. By adjusting the transparency of the layers, the visualization of the raw data together with the summary can be tuned for optimal presentation and interpretation. The app is dubbed PlotsOfData and is available at: https://huygens.science.uva.nl/PlotsOfData/

Sign in / Sign up

Export Citation Format

Share Document