Filters








34,024 Hits in 3.7 sec

Learning-Based Abstractions for Nonlinear Constraint Solving

Sumanth Dathathri, Nikos Arechiga, Sicun Gao, Richard M. Murray
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
We propose a new abstraction refinement procedure based on machine learning to improve the performance of nonlinear constraint solving algorithms on large-scale problems.  ...  The mechanism is capable of producing intermediate symbolic abstractions that are also important for many applications and for understanding the internal structures of hard constraint solving problems.  ...  Burdick for helpful input. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  ... 
doi:10.24963/ijcai.2017/83 dblp:conf/ijcai/DathathriAGM17 fatcat:cv3fh5nmyzfphnevgfwv5k46ju

Learning Neural Networks under Input-Output Specifications [article]

Zain ul Abdeen, He Yin, Vassilis Kekatos, Ming Jin
2022 arXiv   pre-print
To this end, we address the key technical challenge of convexifying the verification condition for neural networks, which is derived by abstracting the nonlinear specifications and activation functions  ...  with quadratic constraints.  ...  SPECIFICATION ANALYSIS FOR A FIXED NN We now briefly review the analysis conducted in [17] based on the framework of quadratic constraints.  ... 
arXiv:2202.11246v1 fatcat:pmfl3xmlarhc3hg47ksc67rpoa

A Distributed Primal Decomposition Scheme for Nonconvex Optimization

Andrea Camisa, Giuseppe Notarstefano
2019 IFAC-PapersOnLine  
Abstract: In this paper, we deal with large-scale nonconvex optimization problems, typically arising in distributed nonlinear optimal control, that must be solved by agents in a network.  ...  Abstract: In this paper, we deal with large-scale nonconvex optimization problems, typically arising in distributed nonlinear optimal control, that must be solved by agents in a network.  ...  Our distributed algorithm enjoys Abstract: In this paper, we deal with large-scale nonconvex optimization problems, typically arising in distributed nonlinear optimal control, that must be solved by agents  ... 
doi:10.1016/j.ifacol.2019.12.174 fatcat:irhzgdtrxbbqlbmz6b3bgxhcze

A Review of Constraint Programming

Poonam Dabas, Vaishali Cooner
2014 International Journal of Computer Applications Technology and Research  
Constraint programming is based on feasibility which means finding a feasible solution rather than optimization which means finding an optimal solution and focuses on the constraints and variables domain  ...  A model could be very hard to solve if it is poorly chosen.  ...  Numerical result for solving large scale nonlinear optimization problems is presented. The performance of each solver is explained easily and predicted based on the characteristics.  ... 
doi:10.7753/ijcatr0307.1001 fatcat:fhokqi424bb57ieuujgunfq5k4

Speeding up problem solving by abstraction: a graph oriented approach

R. Holte, T. Mkadmi, R.M. Zimmer, A.J. McDonald
1996 Artificial Intelligence  
Pearl, Uncovering trees in constraint networks B.J. Grosz and S. Kraus, Collaborative plans for complex group action  ...  Graph search methods for non-order-preserving evaluation functions: applications to job sequencing problems J. Lin, A semantics for reasoning consistently in the presence of inconsistency R.  ...  MacDonald, Speeding up problem solving by abstraction: a graph oriented approach Thl\ paper presents a new perspcctlve on the traditional Al task of problem solving and the techniques of abstraction and  ... 
doi:10.1016/0004-3702(96)81372-5 fatcat:bhdyzgaxhrhepjckt2iqfjp2xm

Exploratory analysis of speedup learning data using expectation maximization

A.M. Segre, G.J. Elkan
1996 Artificial Intelligence  
Pearl, Uncovering trees in constraint networks B.J. Grosz and S. Kraus, Collaborative plans for complex group action  ...  Graph search methods for non-order-preserving evaluation functions: applications to job sequencing problems J. Lin, A semantics for reasoning consistently in the presence of inconsistency R.  ...  MacDonald, Speeding up problem solving by abstraction: a graph oriented approach Thl\ paper presents a new perspcctlve on the traditional Al task of problem solving and the techniques of abstraction and  ... 
doi:10.1016/0004-3702(96)81373-7 fatcat:vznawqt52beqpeedmeens7t7dq

Modular strategic SMT solving with SMT-RAT

Gereon Kremer, Erika Ábrahám
2018 Acta Universitatis Sapientiae: Informatica  
As a distinguishing feature, SMT-RAT provides a set of solving modules and supports their strategic combination.  ...  In this paper we present the latest developments in SMT-RAT, a tool for the automated check of quantifier-free real and integer arithmetic formulas for satisfiability.  ...  Learning the Boolean abstraction of this explanation refines the Boolean abstraction, avoiding Boolean solutions with the same theory conflict in future search.  ... 
doi:10.2478/ausi-2018-0001 fatcat:eedfeixfz5fc5jdph2g6rzrcye

Planning and Operations Research (Dagstuhl Seminar 18071)

J. Christopher Beck, Daniele Magazzeni, Gabriele Röger, Willem-Jan Van Hoeve, Michael Wagner
2018 Dagstuhl Reports  
The seminar brought together researchers in the areas of Artificial Intelligence (AI) Planning, Constraint Programming, and Operations Research.  ...  of artificial intelligence where the emphasis was traditionally more on symbolic and logical search techniques for the intelligent selection and sequencing of actions to achieve a set of goals.  ...  We presented a constraint-based declarative model for MAPF, together with its implementation in Picat, a logic-based programming language.  ... 
doi:10.4230/dagrep.8.2.26 dblp:journals/dagstuhl-reports/BeckMRH18 fatcat:lavt5jfujfarfmtwrpbbxan2oq

Exploring case-Based building design—CADRE

Kefeng Hua, Boi Faltings
1993 Artificial intelligence for engineering design, analysis and manufacturing  
We describe the problems and the ways we either solved or worked around them in the CADRE system.  ...  In our work on CADRE, a case-based building design system, we have encountered seven fundamental problems which we think are common to most case-based design systems.  ...  Conclusion: How far can the potential advantages of case-based design be achieved? Case-based reasoning has been credited for its advantages in solving design problems.  ... 
doi:10.1017/s0890060400000822 fatcat:e7ytoz67mva7lorol7okl37eye

Optimal Path Analysis for Solving Nonlinear Equations with Finite Local Error

Xiaoxiao Ma, Xiaojuan Chen
2022 North atlantic university union: International Journal of Circuits, Systems and Signal Processing  
Because the traditional method of solving nonlinear equations takes a long time, an optimal path analysis method for solving nonlinear equations with limited local error is designed.  ...  Secondly, set the constraints of the objective function, solve the optimal solution of the nonlinear equation under the condition of limited local error, and obtain the optimal path of the nonlinear equation  ...  The vast majority of methods for solving nonlinear equations are iterative methods.  ... 
doi:10.46300/9106.2022.16.13 fatcat:uufdqmtfcnefnmygty5i734g6u

Maximum Likelihood Methods for Inverse Learning of Optimal Controllers

Marcel Menner, Melanie N. Zeilinger
2020 IFAC-PapersOnLine  
Abstract: This paper presents a framework for inverse learning of objective functions for constrained optimal control problems, which is based on the Karush-Kuhn-Tucker (KKT) conditions.  ...  Abstract: This paper presents a framework for inverse learning of objective functions for constrained optimal control problems, which is based on the Karush-Kuhn-Tucker (KKT) conditions.  ...  In simulation, we present learning results for both constrained, linear and nonlinear systems.  ... 
doi:10.1016/j.ifacol.2020.12.1206 fatcat:jyzclz3przdkzozoqesxw5jfma

Page 4238 of Mathematical Reviews Vol. , Issue 2004e [page]

2004 Mathematical Reviews  
The code can also solve second-order conic pro- gramming (SOCP) problems, as well as problems with a mixture of SDP, SOCP and NLP constraints.  ...  Unsupervised learning (i.e., clustering) is done by unconstrained minimization of a non-smooth cluster function, and supervised learning (i.e., classification) then is performed by solving the cluster  ... 

Data-Driven Control of Unknown Systems: A Linear Programming Approach

Alexandros Tanzanakis, John Lygeros
2020 IFAC-PapersOnLine  
Abstract: We consider the problem of discounted optimal state-feedback regulation for general unknown deterministic discrete-time systems.  ...  Abstract: We consider the problem of discounted optimal state-feedback regulation for general unknown deterministic discrete-time systems.  ...  , utilizing a heuristic based on support constraints.  ... 
doi:10.1016/j.ifacol.2020.12.027 fatcat:3psowni2erfflif7h5rq3e64iy

Nonlinear Unmixing of Hyperspectral Data Based on a Linear-Mixture/Nonlinear-Fluctuation Model

Jie Chen, Cédric Richard, Paul Honeine
2013 IEEE Transactions on Signal Processing  
Index Terms-Hyperspectral imaging, multi-kernel learning, nonlinear spectral unmixing, support vector regression.  ...  In this paper, we formulate a new kernel-based paradigm that relies on the assumption that the mixing mechanism can be described by a linear mixture of endmember spectra, with additive nonlinear fluctuations  ...  The abundances are determined by solving an appropriate kernel-based regression problem under constraints. This paper is organized as follows.  ... 
doi:10.1109/tsp.2012.2222390 fatcat:m35fa2xisrcm5o42zhwm4bk7j4

Page 1178 of Mathematical Reviews Vol. , Issue 97B [page]

1997 Mathematical Reviews  
For nonlinear problems, it is shown that the concepts of complexity based on adaptive information are helpful.  ...  binary search, min- imax optimization of unimodal functions and solving of some nonlinear equations.  ... 
« Previous Showing results 1 — 15 out of 34,024 results