Imitation learning, or programming by demonstration (PbD), holds the promise of allowing robots to acquire skills from humans with domain-specific knowledge, who nonetheless are inexperienced at programming robots. We...
详细信息
Imitation learning, or programming by demonstration (PbD), holds the promise of allowing robots to acquire skills from humans with domain-specific knowledge, who nonetheless are inexperienced at programming robots. We have prototyped a real-time, closed-loop system for teaching a humanoid robot to interact with objects in its environment. The system uses nonparametric Bayesian inference to determine an optimal action given a configuration of objects in the world and a desired future configuration. We describe our prototype implementation, show imitation of simple motor acts on a humanoid robot, and discuss extensions to the system
Runtime support for parallelization of scientific programs is needed when some information important for decisions in this process cannot be accurately derived at compile time. This paper describes a project which int...
详细信息
Runtime support for parallelization of scientific programs is needed when some information important for decisions in this process cannot be accurately derived at compile time. This paper describes a project which integrates runtime parallelization with the advanced compile-time parallelization techniques of SUPERB. Besides the description of implementation techniques, language constructs are proposed, providing means for the specification of irregular computations. SUPERB is an interactive SIMD/MIMD parallelizing system for the Suprenum, iPSC/860 and Genesis-P machines. The implementation of the runtime parallelization is based on the Parti procedures developed at ICASE NASA.< >
OSF DCE (Distributed Computing Environment) offers a programming environment that facilitates users to implement clients and servers residing on different workstation. However, developing a DCE client and server appli...
详细信息
OSF DCE (Distributed Computing Environment) offers a programming environment that facilitates users to implement clients and servers residing on different workstation. However, developing a DCE client and server application is not an easy task. DCE offers hundreds of APIs and users will easily get confused with what they need in order to accomplish their mission. This paper presents a design of a Client/Server program construction environment. With this environment users can implement their applications in traditional sequential language. Our system will automatically translate the sequential program into client and server programs. A working prototype has been implemented.
This case study examines the application of grammatical swarm to classification problems, and illustrates the particle swarm algorithms' ability to specify the construction of programs. Each individual particle re...
详细信息
This case study examines the application of grammatical swarm to classification problems, and illustrates the particle swarm algorithms' ability to specify the construction of programs. Each individual particle represents choices of program construction rules, where these rules are specified using a Backus-Naur Form grammar. Two problem instances are tackled, the first a mushroom classification problem, the second a bioinformatics problem that involves the detection of eukaryotic DNA promoter sequences. For the first problem we generate solutions that take the form of conditional statements in a C-like language subset, and for the second problem we generate simple regular expressions. The results demonstrate that it is simple regular expressions. The results demonstrate that it is possible to generate programs using the grammatical swarm technique with a performance similar to the grammatical evolution evolutionary automatic programming approach.
Presents a control and learning architecture for humanoid robots designed for acquiring movement skills in the context of imitation learning. Multiple levels of movement abstraction occur across the hierarchical struc...
详细信息
Presents a control and learning architecture for humanoid robots designed for acquiring movement skills in the context of imitation learning. Multiple levels of movement abstraction occur across the hierarchical structure of the architecture, finally leading to the representation of movement sequences within a probabilistic framework. As its substrate, the framework uses the notion of visuo-motor primitives, modules capable of recognizing as well as executing similar movements. This notion is heavily motivated by the neuroscience evidence for motor primitives and mirror neurons. Experimental results from an implementation of the architecture are presented involving learning and representation of demonstrated movement sequences from synthetic as well as real human movement data.
We show that automata theory can be useful for engineering, developing and documenting multiagent systems. Algorithmization and programming principles for logic control are formulated, regarding algorithms and program...
详细信息
We show that automata theory can be useful for engineering, developing and documenting multiagent systems. Algorithmization and programming principles for logic control are formulated, regarding algorithms and programs as finite automata.
Multilinear algebra kernel performance on modern massively-parallel systems is determined mainly by data movement. However, deriving data movement-optimal distributed schedules for programs with many high-dimensional ...
详细信息
Multilinear algebra kernel performance on modern massively-parallel systems is determined mainly by data movement. However, deriving data movement-optimal distributed schedules for programs with many high-dimensional inputs is a notoriously hard problem. State-of-the-art libraries rely on heuristics and often fall back to suboptimal tensor folding and BLAS calls. We present Deinsum, an automated framework for distributed multilinear algebra computations expressed in Einstein notation, based on rigorous mathematical tools to address this problem. Our framework automatically derives data movement-optimal tiling and generates corresponding distributed schedules, further optimizing the performance of local computations by increasing their arithmetic intensity. To show the benefits of our approach, we test it on two important tensor kernel classes: Matricized Tensor Times Khatri-Rao Products and Tensor Times Matrix chains. We show performance results and scaling on the Piz Daint supercomputer, with up to 19x speedup over state-of-the-art solutions on 512 nodes.
A virtual robot teaching system, consisting of human demonstration and motion-intention analysis in a virtual reality environment, is an advanced form of automatic programming for multi-fingered robots. We propose a n...
详细信息
A virtual robot teaching system, consisting of human demonstration and motion-intention analysis in a virtual reality environment, is an advanced form of automatic programming for multi-fingered robots. We propose a new segmentation method to analyze human motion data. In this method, the human motion data consisting of contact points, grasped force, hand and object position, and the like are segmented into plural primitive motions and the type of task is analyzed based on the sequence of the primitive motions. A trial assembly task using a humanoid robot hand named Gifu Hand III is shown.
A new approach for solving TCAD problems in a multidimensional and multitopological way based on a generic scientific simulation environment is presented. By embedding functional programming, the environment imposes n...
详细信息
A new approach for solving TCAD problems in a multidimensional and multitopological way based on a generic scientific simulation environment is presented. By embedding functional programming, the environment imposes no restrictions on geometry, topology, or discretization schemes. Therewith equations and even complete models can be implemented easily. The generic and functional programming paradigms and the applicability in the area of TCAD are presented
The main purpose of this paper is to propose an incorporating a grammatical evolution (GE) into the genetic algorithm (GA), called GEGA, and apply it to estimate the compressive strength of high-performance concrete (...
详细信息
ISBN:
(纸本)9781424445424
The main purpose of this paper is to propose an incorporating a grammatical evolution (GE) into the genetic algorithm (GA), called GEGA, and apply it to estimate the compressive strength of high-performance concrete (HPC). The GE, an evolutionary programming type system, automatically discovers complex relationships between significant factors and the strength of HPC in a more transparent way to enhance our understanding of the mechanisms. A GA was used afterward with GE to optimize the appropriate function type and associated coefficients using over 1,000 examples for which experimental data were available. The results show that this novel model, GEGA, can obtain a highly nonlinear mathematical equation which outperforms than the traditional multiple regression analysis (RA) with lower estimating errors for predicting the compressive strength of HPC.
暂无评论