scholarly journals Online Speedup Learning for Optimal Planning

2012 ◽  
Vol 44 ◽  
pp. 709-755 ◽  
Author(s):  
C. Domshlak ◽  
E. Karpas ◽  
S. Markovitch
Keyword(s):  
Decision Rule ◽  
Search Space ◽  
Optimal Planning ◽  
Goal State ◽  
World State ◽  
Admissible Heuristics ◽  
State Space Search ◽  
Novel Method ◽  
Planning Task ◽  
The Cost

Domain-independent planning is one of the foundational areas in the field of Artificial Intelligence. A description of a planning task consists of an initial world state, a goal, and a set of actions for modifying the world state. The objective is to find a sequence of actions, that is, a plan, that transforms the initial world state into a goal state. In optimal planning, we are interested in finding not just a plan, but one of the cheapest plans. A prominent approach to optimal planning these days is heuristic state-space search, guided by admissible heuristic functions. Numerous admissible heuristics have been developed, each with its own strengths and weaknesses, and it is well known that there is no single "best'' heuristic for optimal planning in general. Thus, which heuristic to choose for a given planning task is a difficult question. This difficulty can be avoided by combining several heuristics, but that requires computing numerous heuristic estimates at each state, and the tradeoff between the time spent doing so and the time saved by the combined advantages of the different heuristics might be high. We present a novel method that reduces the cost of combining admissible heuristics for optimal planning, while maintaining its benefits. Using an idealized search space model, we formulate a decision rule for choosing the best heuristic to compute at each state. We then present an active online learning approach for learning a classifier with that decision rule as the target concept, and employ the learned classifier to decide which heuristic to compute at each state. We evaluate this technique empirically, and show that it substantially outperforms the standard method for combining several heuristics via their pointwise maximum.

scholarly journals Efficient Guiding Towards Cost-Optimality in UPPAAL

2001 ◽  
Vol 8 (4) ◽  
Author(s):  
Gerd Behrmann ◽  
Ansgar Fehnker ◽  
Thomas S. Hune ◽  
Kim G. Larsen ◽  
Paul Pettersson ◽  
...  
Keyword(s):  
Timed Automata ◽  
Minimum Cost ◽  
Search Space ◽  
Shortest Path Algorithm ◽  
Goal State ◽  
Verification Tool ◽  
Search Order ◽  
Efficient Representation ◽  
The Cost ◽  
Cost Optimality

<p>In this paper we present an algorithm for efficiently computing<br /> the minimum cost of reaching a goal state in the model of Uniformly<br />Priced Timed Automata (UPTA). This model can be seen as a submodel<br />of the recently suggested model of linearly priced timed automata, which<br />extends timed automata with prices on both locations and transitions.<br />The presented algorithm is based on a symbolic semantics of UTPA, and<br />an efficient representation and operations based on difference bound <br />matrices. In analogy with Dijkstra's shortest path algorithm, we show that<br />the search order of the algorithm can be chosen such that the number of<br />symbolic states explored by the algorithm is optimal, in the sense that<br />the number of explored states can not be reduced by any other search<br />order based on the cost of states. We also present a number of techniques<br />inspired by branch-and-bound algorithms which can be used for limiting<br />the search space and for quickly finding near-optimal solutions.<br />The algorithm has been implemented in the verification tool Uppaal.<br />When applied on a number of experiments the presented techniques <br />reduced the explored state-space with up to 90%.</p>

scholarly journals Deep Learning for Cost-Optimal Planning: Task-Dependent Planner Selection

2019 ◽  
Vol 33 ◽  
pp. 7715-7723 ◽  
Author(s):  
Silvan Sievers ◽  
Michael Katz ◽  
Shirin Sohrabi ◽  
Horst Samulowitz ◽  
Patrick Ferber
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Classification Task ◽  
Automatic Extraction ◽  
Optimal Planning ◽  
Statistical Features ◽  
Planning Task ◽  
International Planning ◽  
The Cost ◽  
Online Portfolio

As classical planning is known to be computationally hard, no single planner is expected to work well across many planning domains. One solution to this problem is to use online portfolio planners that select a planner for a given task. These portfolios perform a classification task, a well-known and wellresearched task in the field of machine learning. The classification is usually performed using a representation of planning tasks with a collection of hand-crafted statistical features. Recent techniques in machine learning that are based on automatic extraction of features have not been employed yet due to the lack of suitable representations of planning tasks.In this work, we alleviate this barrier. We suggest representing planning tasks by images, allowing to exploit arguably one of the most commonly used and best developed techniques in deep learning. We explore some of the questions that inevitably rise when applying such a technique, and present various ways of building practically useful online portfoliobased planners. An evidence of the usefulness of our proposed technique is a planner that won the cost-optimal track of the International Planning Competition 2018.

scholarly journals Implicit Abstraction Heuristics

2010 ◽  
Vol 39 ◽  
pp. 51-126 ◽  
Author(s):  
M. Katz ◽  
C. Domshlak
Keyword(s):  
Heuristic Search ◽  
State Of The Art ◽  
Empirical Evaluation ◽  
The State ◽  
Optimal Planning ◽  
Heuristic Function ◽  
Admissible Heuristics ◽  
State Space Search ◽  
Abstract Spaces ◽  
Exponential Size

State-space search with explicit abstraction heuristics is at the state of the art of cost-optimal planning. These heuristics are inherently limited, nonetheless, because the size of the abstract space must be bounded by some, even if a very large, constant. Targeting this shortcoming, we introduce the notion of (additive) implicit abstractions, in which the planning task is abstracted by instances of tractable fragments of optimal planning. We then introduce a concrete setting of this framework, called fork-decomposition, that is based on two novel fragments of tractable cost-optimal planning. The induced admissible heuristics are then studied formally and empirically. This study testifies for the accuracy of the fork decomposition heuristics, yet our empirical evaluation also stresses the tradeoff between their accuracy and the runtime complexity of computing them. Indeed, some of the power of the explicit abstraction heuristics comes from precomputing the heuristic function offline and then determining h(s) for each evaluated state s by a very fast lookup in a ``database.'' By contrast, while fork-decomposition heuristics can be calculated in polynomial time, computing them is far from being fast. To address this problem, we show that the time-per-node complexity bottleneck of the fork-decomposition heuristics can be successfully overcome. We demonstrate that an equivalent of the explicit abstraction notion of a ``database'' exists for the fork-decomposition abstractions as well, despite their exponential-size abstract spaces. We then verify empirically that heuristic search with the ``databased" fork-decomposition heuristics favorably competes with the state of the art of cost-optimal planning.

scholarly journals Cost Analysis of a Novel Method for Ecological Compensation—A Study of the Translocation of Dead Wood

2021 ◽  
Vol 13 (11) ◽  
pp. 6075
Author(s):  
Ola Lindroos ◽  
Malin Söderlind ◽  
Joel Jensen ◽  
Joakim Hjältén
Keyword(s):  
Cost Efficiency ◽  
Dead Wood ◽  
Road Transportation ◽  
Ecological Value ◽  
Ecological Compensation ◽  
Work Time ◽  
Delivery Times ◽  
Novel Method ◽  
Cost Efficient ◽  
The Cost

Translocation of dead wood is a novel method for ecological compensation and restoration that could, potentially, provide a new important tool for biodiversity conservation. With this method, substrates that normally have long delivery times are instantly created in a compensation area, and ideally many of the associated dead wood dwelling organisms are translocated together with the substrates. However, to a large extent, there is a lack of knowledge about the cost efficiency of different methods of ecological compensation. Therefore, the costs for different parts of a translocation process and its dependency on some influencing factors were studied. The observed cost was 465 SEK per translocated log for the actual compensation measure, with an additional 349 SEK/log for work to enable evaluation of the translocation’s ecological results. Based on time studies, models were developed to predict required work time and costs for different transportation distances and load sizes. Those models indicated that short extraction and insertion distances for logs should be prioritized over road transportation distances to minimize costs. They also highlighted a trade-off between costs and time until a given ecological value is reached in the compensation area. The methodology used can contribute to more cost-efficient operations and, by doing so, increase the use of ecological compensation and the benefits from a given input.

scholarly journals Embedding Undersampling Rotation Forest for Imbalanced Problem

2018 ◽  
Vol 2018 ◽  
pp. 1-15 ◽  
Author(s):  
Huaping Guo ◽  
Xiaoyu Diao ◽  
Hongbing Liu
Keyword(s):  
Imbalanced Data ◽  
Feature Space ◽  
Original Data ◽  
Training Set ◽  
Data Set ◽  
Minority Class ◽  
Rotation Forest ◽  
Novel Method ◽  
Individual Classifier ◽  
The Cost

Rotation Forest is an ensemble learning approach achieving better performance comparing to Bagging and Boosting through building accurate and diverse classifiers using rotated feature space. However, like other conventional classifiers, Rotation Forest does not work well on the imbalanced data which are characterized as having much less examples of one class (minority class) than the other (majority class), and the cost of misclassifying minority class examples is often much more expensive than the contrary cases. This paper proposes a novel method called Embedding Undersampling Rotation Forest (EURF) to handle this problem (1) sampling subsets from the majority class and learning a projection matrix from each subset and (2) obtaining training sets by projecting re-undersampling subsets of the original data set to new spaces defined by the matrices and constructing an individual classifier from each training set. For the first method, undersampling is to force the rotation matrix to better capture the features of the minority class without harming the diversity between individual classifiers. With respect to the second method, the undersampling technique aims to improve the performance of individual classifiers on the minority class. The experimental results show that EURF achieves significantly better performance comparing to other state-of-the-art methods.

scholarly journals Solving Delete Free Planning with Relaxed Decision Diagram Based Heuristics

2020 ◽  
Vol 67 ◽  
pp. 607-651
Author(s):  
Margarita Paz Castro ◽  
Chiara Piacentini ◽  
Andre Augusto Cire ◽  
J. Christopher Beck
Keyword(s):  
Linear Programming ◽  
Empirical Analysis ◽  
State Of The Art ◽  
The State ◽  
Decision Diagrams ◽  
Binary Decision ◽  
Admissible Heuristics ◽  
The Cost ◽  
Construction Algorithms ◽  
Diagram Representation

We investigate the use of relaxed decision diagrams (DDs) for computing admissible heuristics for the cost-optimal delete-free planning (DFP) problem. Our main contributions are the introduction of two novel DD encodings for a DFP task: a multivalued decision diagram that includes the sequencing aspect of the problem and a binary decision diagram representation of its sequential relaxation. We present construction algorithms for each DD that leverage these different perspectives of the DFP task and provide theoretical and empirical analyses of the associated heuristics. We further show that relaxed DDs can be used beyond heuristic computation to extract delete-free plans, find action landmarks, and identify redundant actions. Our empirical analysis shows that while DD-based heuristics trail the state of the art, even small relaxed DDs are competitive with the linear programming heuristic for the DFP task, thus, revealing novel ways of designing admissible heuristics.

scholarly journals Best-first Enumeration Based on Bounding Conflicts, and its Application to Large-scale Hybrid Estimation (Extended Abstract)

2020 ◽  
Author(s):  
Eric Timmons ◽  
Brian C. Williams
Keyword(s):  
Large Scale ◽  
Search Space ◽  
Estimation Methods ◽  
Directed Search ◽  
Hybrid Estimation ◽  
Continuous State ◽  
Order Of Magnitude ◽  
State Models ◽  
The Cost ◽  
Belief States

State estimation methods based on hybrid discrete and continuous state models have emerged as a method of precisely computing belief states for real world systems, however they have difficulty scaling to systems with more than a handful of components. Classical, consistency based diagnosis methods scale to this level by combining best-first enumeration and conflict-directed search. While best-first methods have been developed for hybrid estimation, conflict-directed methods have thus far been elusive as conflicts summarize constraint violations, but probabilistic hybrid estimation is relatively unconstrained. In this paper we present an approach (A*BC) that unifies best-first enumeration and conflict-directed search in relatively unconstrained problems through the concept of "bounding" conflicts, an extension of conflicts that represent tighter bounds on the cost of regions of the search space. Experiments show that an A*BC powered state estimator produces estimates up to an order of magnitude faster than the current state of the art, particularly on large systems.

Design of Fazl-Tash novel method for sustainable resilient comprehensive supplier selection problem

Kybernetes ◽  
2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Hamed Fazlollahtabar ◽  
Navid Kazemitash
Keyword(s):  
Supplier Selection ◽  
Huge Number ◽  
Simple Method ◽  
Content Type ◽  
Novel Method ◽  
Rank And Select ◽  
The Cost ◽  
Supplier Selection Problem ◽  
Industrial Company

Purpose However, due to the huge number of studies and on the other hand to be new and creative, the represented models and methods – as the two main parts of this field – have been got more complicated, which consequently have been turned into unpractical research studies for the realistic situations. Therefore, the purpose of this study is the representation of a novel and simple method to deal with the aforementioned gap. Design/methodology/approach To this end, Fazl-Tash method have been proposed, in which a thorough and complete model including 114 criteria and a simple technique to rank and select the best supplier have been presented. Sustainability and resiliency are considered in collecting criteria effective on supplier selection. Findings The method was carried out in a case study in an industrial company. The efficiency of the proposed method is evaluated in comparison with other conventional approaches. Originality/value As selecting the supplier plays a crucial role to bring some important advantages for companies, such as coping with the cost and time problems and influencing the majority of contemporary markets’ requirements, in recent years, there have been representing more effective studies in the supplier selection literature.

A hybrid framework to model factors affecting construction labour productivity

2019 ◽  
Vol 24 (3) ◽  
pp. 630-654 ◽  
Author(s):  
Majid Parchami Jalal ◽  
Shahab Shoar
Keyword(s):  
Construction Industry ◽  
Labour Productivity ◽  
Content Type ◽  
Factors Affecting ◽  
Schedule Delay ◽  
Factors Influencing ◽  
Hybrid Framework ◽  
Novel Method ◽  
The Cost ◽  
The Impact

Purpose This paper aims to model different causal relations among factors interacting with labour productivity in order to recognize the most important factors influencing and influenced by it. Design/methodology/approach Top 60 factors affecting labour productivity were determined and grouped into 5 major groups by reviewing previous research and interviewing relevant experts. The interactions of factors were modelled using system dynamics (SD) approach. The resulting causal loop diagrams obtained from SD were then applied for identifying the most crucial factors influencing and influenced by labour productivity through the decision-making trial and evaluation laboratory (DEMATEL) method. The impact of factors on each other was finally determined based on the opinions of 63 experts selected from the Iranian construction industry. Findings The results indicated that factors such as fatigue, lack of labour motivation and lack of skill are the most influencing, and factors such as schedule delay and inflation in the cost of execution are the most influenced by labour productivity. In the end, a set of recommendations to improve construction labour productivity was also presented. Originality/value The main contribution of the study is proposing a novel method which is capable of providing insights into how causes and effects of construction labour productivity are interrelated. Furthermore, the proposed method makes this study distinct from previous research in the light of prioritizing factors and offering recommendations according to the interrelationships among factors.

scholarly journals Sine-Cosine Optimization-Based Bijective Substitution-Boxes Construction Using Enhanced Dynamics of Chaotic Map

Complexity ◽  
2018 ◽  
Vol 2018 ◽  
pp. 1-16 ◽  
Author(s):  
Amer Awad Alzaidi ◽  
Musheer Ahmad ◽  
Hussam S. Ahmed ◽  
Eesa Al Solami
Keyword(s):  
Fitness Function ◽  
Chaotic Map ◽  
Search Space ◽  
Initial Population ◽  
Substitution Boxes ◽  
Novel Method ◽  
Sine Cosine Algorithm ◽  
And Performance ◽  
Optimization Mechanism ◽  
Independence Criterion

This paper proposes a novel method of constructing strong substitution-boxes (S-boxes) of order n (4 ≤ n ≤ 8) based on a recent optimization algorithm known as sine-cosine algorithm (SCA). The paper also proposes a new 1D chaotic map, which owns enhanced dynamics compared to conventional chaotic map, for generating initial population of S-boxes and facilitating the optimization mechanism of SCA. The proposed method applies the SCA with enhanced chaotic map to explore and exploit the search space for obtaining optimized S-boxes on the basis of maximization of nonlinearity as fitness function. The S-box construction involves three phases such as initialization of population, optimization, and adjustment. The simulation and performance analyses are done using standard measures of nonlinearity, strict avalanche criterion, bits independence criterion, differential uniformity, linear approximation probability, and autocorrelation function. The obtained experimental results are compared with some immediate optimization-based and other S-boxes to show the strength of proposed method for constructing bijective S-boxes of salient cryptographic features.

Sign in / Sign up

Export Citation Format

Share Document