Rapid Concept Learning for Mobile Robots

Concept learning in robotics is an extremely challenging problem: sensory data is often high dimensional, and noisy due to specularities and other irregularities. In this paper, we investigate two general strategies to speed up learning, based on spatial decomposition of the sensory representation, and simultaneous learning of multiple classes using a shared structure. We study two concept learning scenarios: a hallway navigation problem, where the robot has to induce features such as opening or wall. The second task is recycling, where the robot has to learn to recognize objects, such as a trash can. We use a common underlying function approximator in both studies in the form of a feedforward neural network, with several hundred input units and multiple output units. Despite the high degree of freedom afforded by such an approximator, we show the two strategies provide sufficient bias to achieve rapid learning. We provide detailed experimental studies on an actual mobile robot called PAVLOV to illustrate the effectiveness of this approach.     

Enhancing the performance of an agent-based manufacturing system through learning and forecasting

Agent-based technology has been identified as an important approach for developing next generation manufacturing systems. One of the key techniques needed for implementing such advanced systems will be learning. This paper first discusses learning issues in agent-based manufacturing systems and reviews related approaches, then describes how to enhance the performance of an agent-based manufacturing system through learning from history (based on distributed case-based learning and reasoning) and learning from the future (through system forecasting simulation). Learning from history is used to enhance coordination capabilities by minimizing communication and processing overheads. Learning from the future is used to adjust promissory schedules through forecasting simulation, by taking into account the shop floor interactions, production and transportation time. Detailed learning and reasoning mechanisms are described and partial experimental results are presented.     

The role of “craft language” in learning “Waza”

The role of craft language in the process of teaching (learning) Waza (skill) will be discussed from the perspective of human intelligence.It may be said that the ultimate goal of learning Waza in any Japanese traditional performance is not the perfect reproduction of the teaching (learning) process of Waza. In fact, a special metaphorical language (craft language) is used, which has the effect of encouraging the learner to activate his creative imagination. It is through this activity that the he learns his own habitus (Kata).It is suggested that, in considering the difference of function between natural human intelligence and artificial intelligence, attention should be paid to the imaginative activity of the learner as being an essential factor for mastering Kata.This article is a modified English version of Chapter 5 of my bookWaza kara shiru (Learning from Skill), Tokyo University Press, 1987, pp. 93–105.     

Modelling structures in generic space, a condition for adaptiveness of monitoring cognitive agent

The adaptiveness of agents is one of the basic conditions for the autonomy. This paper describes an approach of adaptiveness forMonitoring Cognitive Agents based on the notion of generic spaces. This notion allows the definition of virtual generic processes so that any particular actual process is then a simple configuration of the generic process, that is to say a set of values of parameters. Consequently, generic domain ontology containing the generic knowledge for solving problems concerning the generic process can be developed. This lead to the design of Generic Monitoring Cognitive Agent, a class of agent in which the whole knowledge corpus is generic. In other words, modeling a process within a generic space becomes configuring a generic process and adaptiveness becomes genericity, that is to say independence regarding technology. In this paper, we present an application of this approach on Sachem, a Generic Monitoring Cognitive Agent designed in order to help the operators in operating a blast furnace. Specifically, the NeuroGaz module of Sachem will be used to present the notion of a generic blast furnace. The adaptiveness of Sachem can then be noted through the low cost of the deployment of a Sachem instance on different blast furnaces and the ability of NeuroGaz in solving problem and learning from various top gas instrumentation.     

Logic Based Abstractions of Real-Time Systems

When verifying concurrent systems described by transition systems, state explosion is one of the most serious problems. If quantitative temporal information (expressed by clock ticks) is considered, state explosion is even more serious. We present a notion of abstraction of transition systems, where the abstraction is driven by the formulae of a quantitative temporal logic, called qu-mu-calculus, defined in the paper. The abstraction is based on a notion of bisimulation equivalence, called , n-equivalence, where is a set of actions and n is a natural number. It is proved that two transition systems are , n-equivalent iff they give the same truth value to all qu-mu-calculus formulae such that the actions occurring in the modal operators are contained in , and with time constraints whose values are less than or equal to n. We present a non-standard (abstract) semantics for a timed process algebra able to produce reduced transition systems for checking formulae. The abstract semantics, parametric with respect to a set of actions and a natural number n, produces a reduced transition system , n-equivalent to the standard one. A transformational method is also defined, by means of which it is possible to syntactically transform a program into a smaller one, still preserving , n-equivalence.     

Multi-time scale zero-sum differential games with perfect state measurements

We obtain necessary and sufficient conditions for the existence of approximate saddle-point solutions in linear-quadratic zero-sum differential games when the state dynamics are defined on multiple (three) time scales. These different time scales are characterized in terms of two small positive parameters ₁, and ₂, and the terminology approximate saddle-point solution is used to refer to saddle-point policies that do not depend on ₁ and ₂, while providing cost levels withinO(₁) of the full-order game. It is shown in the paper that, under perfect state measurements, the original game can be decomposed into three subgames-slow, fast and fastest, the composite saddle-point solution of which make up the approximate saddle-point solution of the original game. Specifically, for the minimizing player, it is necessary to use a composite policy that uses the solutions of all three subgames, whereas for the maximizing player, it is sufficient to use a slow policy. In the finite-horizon case this slow policy could be a feedback policy, whereas in the infinite-horizon case it has to be chosen as an open-loop policy that is generated from the solution and dynamics of the slow subgame. These results have direct applications in theH -optimal control of three-time scale singularly perturbed linear systems under perfect state measurements.Research supported in part by the National Science Foundation under Grant ECS 91-13153, and in part by the U.S. Department of Energy under Grant DE-FG-02-88-ER-13939. This paper is dedicated to the memory of John Breakwell.     

Toward Third Generation Threaded Discussions for Mobile Learning: Opportunities and Challenges for Ubiquitous Collaborative Environments

The mobile communication revolution has led to pervasive connectedness—as evidenced by the explosive growth of instant messaging in the home, and more recently, the enterprise–and, together with the convergence of mobile computing, provides a basis for extending collaborative environments toward truly ubiquitous immersion. Leveraging the true anytime/anywhere access afforded by mobile computing, it becomes possible to develop applications that not only are capable of responding to users whenever/wherever, on demand, but that also may actively seek out and engage users when the need arises. Thus, immersive environments need no longer be thought of strictly in terms of physical immersion with clearly discernable enter and exit events, but rather they may be extended, through mobile-enabled computing, toward ubiquity in terms of both time and space. Based on Media Synchronicity Theory, potential benefits are envisioned, particularly in the case of collaborative learning environments, from shortened response cycles and increased real time interaction opportunities. At the same time, a number of challenging issues must be addressed in designing such an environment to ensure user acceptance and to maximize realization of the potential. Third Generation (3G) Threaded Discussion has been conceptualized as an environment, well suited to mobile learning (m-learning) that could leverage mobile-enabled ubiquity to achieve a degree of extended immersion and thereby accrue the associated collaboration benefits. Exploring this conceptualization serves to help surface both the opportunities and the challenges associated with such environments and to identify promising design approaches, such as the use of intelligent agents.This revised version was published online in March 2005 with corrections to the cover date     

Parallel geometric algorithms on a mesh-connected computer

We show that a number of geometric problems can be solved on a n × n mesh-connected computer (MCC) inO(n) time, which is optimal to within a constant factor, since a nontrivial data movement on an MCC requires (n) time. The problems studied here include multipoint location, planar point location, trapezoidal decomposition, intersection detection, intersection of two convex polygons, Voronoi diagram, the largest empty circle, the smallest enclosing circle, etc. TheO(n) algorithms for all of the above problems are based on the classical divide-and-conquer problem-solving strategy.This work was supported in part by the National Science Foundation under Grant DCR 8420814. A preliminary version was presented in the 1987 FJCC, Dallas, TX.     

Low-Rank Approximation of Integral Operators by Interpolation

A central component of the analysis of panel clustering techniques for the approximation of integral operators is the so-called -admissibility condition min {diam(),diam()} 2dist(,) that ensures that the kernel function is approximated only on those parts of the domain that are far from the singularity. Typical techniques based on a Taylor expansion of the kernel function require a subdomain to be far enough from the singularity such that the parameter has to be smaller than a given constant depending on properties of the kernel function. In this paper, we demonstrate that any is sufficient if interpolation instead of Taylor expansionisused for the kernel approximation, which paves the way for grey-box panel clustering algorithms.     

Information and pragmatic value-adding: Language games and information science

This paper discusses terms which are of mutual importance to the fields of information science and computer science. Specifically we discuss the notions of information and knowledge: their interrelationships as well as their differences, and the concept of value-adding. Concrete examples are used in the discussion.Rainer Kuhlen is professor of Information Science at the University of Konstanz.     

On the Influence of Scale Selection on Feature Detection for the Case of Linelike Structures

A method to automatically select locally appropriate scales for feature detection, proposed by Lindeberg (1993b, 1998), involves choosing a so-called -parameter. The implications of the choice of -parameter are studied and it is demonstrated that different values of can lead to qualitatively different features being detected. As an example the range of -values is determined such that a second derivative of Gaussian filter kernel detects ridges but not edges. Some results of this relatively simple ridge detector are shown for two-dimensional images.     

A Note on the Ohya-Masuda Quantum Algorithm

We analyze the Ohya-Masuda quantum algorithm that solves the so-called satisfiability problem, which is an NP-complete problem of the complexity theory. We distinguish three steps in the algorithm, and analyze the second step, in which a coherent superposition of states (a pure state) transforms into an incoherent mixture presented by a density matrix. We show that, if nonideal (in analogy with nonideal quantum measurement), this transformation can make the algorithm to fail in some cases. On this basis we give some general notions on the physical implementation of the Ohya-Masuda algorithm.     

Informatik als Grundbildung

Im Teil II Bildung von Elementarsätzen wird der Teil Namengebung und Kennzeichnung vorbereitet. Seine Überschrift könnte auch lauten: Was sind Nominatoren und wie kommen sie zustande? Teil V ist wie seine Vorgänger Teil III Gleichheit und Abstraktion und Teil IV Objektsprache/Metasprache als eine Erweiterung einer rationalen Grammatik aufzufassen. Der noch folgende Teil VI Logik und Geltungsicherung von Behauptungen schließt die Serie Informatik als Grundbildung ab.     

From Conditional Events to Conditional Measures: A New Axiomatic Approach

Our starting point is a definition of conditional event EH which differs from many seemingly similar ones adopted in the relevant literature since 1935, starting with de Finetti. In fact, if we do not assign the same third value u (undetermined) to all conditional events, but make it depend on EH, it turns out that this function t(EH) can be taken as a general conditional uncertainty measure, and we get (through a suitable – in a sense, compulsory – choice of the relevant operations among conditional events) the natural axioms for many different (besides probability) conditional measures.     

Decomposition of Methods and Parallelization of Algorithms for Solving Aerodynamics and Physical Gas Dynamics Problems: Computer System Potok-3

This paper is aimed at developing modern computer technologies and methods of concurrent programming designed to improve the efficiency of solving fundamental scientific and applied problems of numerical simulation in aerodynamics and physical gas dynamics. Theoretical issues concerning the design of aircraft involving large amounts of computations are also examined. This paper is focused on theoretical issues of paralleling available methods and algorithms for solving complex integro-differential systems of equations. A number of paralleling methods are considered based on various types of decomposition of the original problem into several subproblems that can be solved concurrently (decomposition by physicomathematical processes of the technological chain of computations and by the corresponding processes that can be performed concurrently (horizontal paralleling), and decomposition by the computational geometric domains and by the corresponding processes that can be performed concurrently (vertical paralleling)). The efficiency of these methods is compared and the optimal paralleling method is found. The results of the theoretical study are used in developing the third-generation package of programs for solving aerodynamics and gas dynamics problems called Potok-3.     

Transformation of programs for fault-tolerance

In this paper we describe how a program constructed for afault-free system can be transformed into afault-tolerant program for execution on a system which is susceptible to failures. A program is described by a set of atomic actions which perform transformations from states to states. We assume that a fault environment is represented by a programF. Interference by the fault environmentF on the execution of a programP can then be described as afault-transformation which transformsP into a program (P). This is proved to be equivalent to the programPP _F , whereP _F is derived fromP andF, and defines the union of the sets of actions ofP andF _P . A recovery transformation transformsP into a program (P) =PR by adding a set ofrecovery actions R, called arecovery program. If the system isfailstop and faults do not affect recovery actions, we have ((P))=(P)R=PP _F R We illustrate this approach to fault-tolerant programming by considering the problem of designing a protocol that guarantees reliable communication from a sender to a receiver in spite of faults in the communication channel between them.     

Choosing reference classes and building provisional models

A well-known problem in default logic is the ability of naive reasoners to explain bothg and ¬g from a set of observations. This problem is treated in at least two different ways within that camp.One approach is examination of the various explanations and choosing among them on the basis of various explanation comparators. A typical comparator is choosing the explanation that depends on the most specific observation, similar to the notion of narrowest reference class.Others examine default extensions of the observations and choose whatever is true in any extension, or what is true in all extensions or what is true in preferred extensions. Default extensions are sometimes thought of as acceptable models of the world that are discarded as more knowledge becomes available.We argue that the notions of specificity and extension lack clear semantics. Furthermore, we show that the problems these ideas were supposed to solve can be handled easily within a probabilistic framework.     

Adding a temporal dimension to a logic system

We introduce a methodology whereby an arbitrary logic system L can be enriched with temporal features to create a new system T(L). The new system is constructed by combining L with a pure propositional temporal logic T (such as linear temporal logic with Since and Until) in a special way. We refer to this method as adding a temporal dimension to L or just temporalising L. We show that the logic system T(L) preserves several properties of the original temporal logic like soundness, completeness, decidability, conservativeness and separation over linear flows of time. We then focus on the temporalisation of first-order logic, and a comparison is make with other first-order approaches to the handling of time.     

Recognition of handwritten musical notes by a modified Neocognitron

A neural network for recognition of handwritten musical notes, based on the well-known Neocognitron model, is described. The Neocognitron has been used for the what pathway (symbol recognition), while contextual knowledge has been applied for the where (symbol placement). This way, we benefit from dividing the process for dealing with this complicated recognition task. Also, different degrees of intrusiveness in learning have been incorporated in the same network: More intrusive supervised learning has been implemented in the lower neuron layers and less intrusive in the upper one. This way, the network adapts itself to the handwriting of the user. The network consists of a 13×49 input layer and three pairs of simple and complex neuron layers. It has been trained to recognize 20 symbols of unconnected notes on a musical staff and was tested with a set of unlearned input notes. Its recognition rate for the individual unseen notes was up to 93%, averaging 80% for all categories. These preliminary results indicate that a modified Neocognitron could be a good candidate for identification of handwritten musical notes.     

Mathematical systems theory: Causality

The concept of state is studied in a new set-theoretic formalism for systems theory. Starting with the notion of a time system as a set of ordered pairs of abstract time functions, the concepts of (i) non-anticipation and (ii) causality are introduced. It is proved that the class of causal systems (those possessing a set of states and functional state transitions) is precisely the class of non-anticipatory systems. It is shown that every causal system has a series decomposition consisting of a transition system followed by a static system. It is proved that a state set for a causal system is always constructible using a class of natural partitions of the system input set. This latter construction generalizes the result known for certain functional discrete systems to a much more general situation.The author gratefully acknowledges the contribution of M. D. Mesarovi to this paper. The research has been supported in part by the Office of Naval Research (Contract No. Nonr 1141(12)).