In this talk I shall try and argue that appropriate techniques based on formalisms based on modal... more In this talk I shall try and argue that appropriate techniques based on formalisms based on modal logic provide a powerful tool for the specification and verification of multiagent systems. The talk will consist of two parts. In the first part the idea of specifications of multiagent systems by means of temporal epistemic logic will be put forward and the main results presented. In the second, attention will be given to the problem of verifying that a multiagent system follows said specifications; the main techniques for verification by model checking (via OBDD and SAT) will be introduced and demonstrated.
This paper describes first steps towards the formal specification and verification of Distributed... more This paper describes first steps towards the formal specification and verification of Distributed Artificial Intelligence (DAI) systems, through the use of temporal belief logics. The paper first describes Concurrent METATEM, a programming language for DAI, and then develops a logic that may be used to reason about Concurrent METATEM systems. The utility of this logic for specifying and verifying Concurrent METATEM systems is demonstrated through a number of examples. The paper concludes with a brief discussion of the wider implications of the work, and in particular on the use of similar logics for reasoning about DAI systems in general.
Copyright is held by the author/owner(s). Submission for Review Article ACM Communications. these... more Copyright is held by the author/owner(s). Submission for Review Article ACM Communications. these questions, we will first recap autonomous systems (in Section 2) and formal verification (in Section 3).
The purpose of this paper is twofold: (i) to illustrate and re-emphasize the use of CONCURRENT ME... more The purpose of this paper is twofold: (i) to illustrate and re-emphasize the use of CONCURRENT METATEM, a programming language based on executable temporal logic, as a viable framework in which to develop Distributed Artificial Intelligence (DAI) applications; and (ii) to present a specific example of a cooperative protocol, give an outline of its implementation in CONCURRENT METATEM, and show how properties of the protocol can be established through formal proof.
METATEM is a simple programming language based on the direct execution of temporal logic statemen... more METATEM is a simple programming language based on the direct execution of temporal logic statements. It was introduced through a number of papers [35,2,3] culminating in a book collecting together work on the basic temporal language [5]. However, since that time, there has been a programme of research, carried out over a number of years, extending, adapting and applying the basic approach. In particular, much of the research has concerned the development of descendents of METATEM for describing and implementing complex multi-agent systems. Thus, while there are a number of other approaches to executing temporal statements [32,18], we will concentrate on this one particular approach and will describe the developments over the last 15 years. The structure of this article mirrors the research developments in that the path through these developments is not linear. The diagram below gives a pictorial explanation of the sections that follow.
The 1998 Workshop of the UK Special Interest Group on Multi-Agent Systems was held in Manchester ... more The 1998 Workshop of the UK Special Interest Group on Multi-Agent Systems was held in Manchester in December, chaired and organised by Michael Fisher of Manchester Metropolitan University, continuing the series of focussed and constructive meetings in this field. After two very successful workshops on the Foundations of Multi-Agent Systems at the University of Warwick in 1996 (Luck, 1997; Doran et al., 1997; d'Inverno et al., 1997; Fisher et al., 1997) and 1997 (Luck et al., 1998; Aylett et al., 1998; Binmore et al., 1998), the ...
Interest in agent-oriented technology continues to grow, both at a theoretical level and a practi... more Interest in agent-oriented technology continues to grow, both at a theoretical level and a practical level, with the UK maintaining a strong representation in the area in both academic institutions and commercial organisations. In December 1997, the Second UK Workshop on Foundations of Multi-Agent Systems (FoMAS'97), held at the University of Warwick, built on the success of FoMAS'96 a year earlier (Luck, 1997; Doran et al., 1997; d'Inverno et al., 1997; Fisher et al., 1997) in seeking to provide a forum for academics and industrialists within the UK to present and discuss current directions in research and applications development. Supported by the EPSRC and HP Labs, FoMAS'97 was expanded to two days with invited presentations from Professor Ken Binmore of UCL, and Professor Cristiano Castelfranchi of the Italian National Research Council's Institute of Psychology, in addition to paper presentations and panel discussions. The aim again was to provide an opportunity for promoting and supporting activity in the research and development of multi-agent systems. The two panels, which were concerned with Agent Systems and Applications, and Agent Rationality, comprising leading academics and industrialists, each generated stimulating and sometimes heated debates. Reports on these panel discussions follow separately. In this report, we summarise the other contributions to the workshop through paper presentations and invited talks, which cover a wide range of relevant topics. The structure of the report reflects the organisation of the workshop. The workshop began with an invited talk from Castelfranchi, who focussed on foundations of multi-agent systems with a social and psychological perspective. His starting point is the new AI of the '90s in which an important stream is artificial social intelligence. The main claims of the talk can be summarised as follows. First, the real foundation of all sociality (cooperation, competition, groups, organisation, etc.) is individual social action and mind. One cannot reduce or connect action at the collective level to action at the individual level unless one passes through the social character of individual action. Second, important levels of coordination and cooperation necessarily require minds and cognitive agents (beliefs, desires, intentions, etc.). However, cognition, communication and agreement are not enough for modelling and implementing cooperation: emergent pre-cognitive structures and constraints should be formalised, and emergent forms of cooperation are also needed among planning and deliberative agents. The final claim was that we are going towards a synthetic paradigm in AI and Cognitive Science, reconciling situatedness and plans, reactivity and mental representations, cognition, emergence and self-organisation. £ This report summarises the paper presentations at the Second UK Workshop on Foundations of Multi-Agent Systems.
As computer scientists, our goals are motivated by the desire to improve computer systems in some... more As computer scientists, our goals are motivated by the desire to improve computer systems in some way: making them easier to design and implement, more robust and less prone to error, easier to use, faster, cheaper, and so on. In the field of multi-agent systems, our goal is to build systems capable of flexible autonomous decision making, with societies of such systems cooperating with one-another. There is a lot of formal theory in the area but it is often not obvious what such theories should represent and what role the theory is ...
The mechanism of formation of the in-chain, unsaturated fatty acid metabolite, ⌬ 3-valproic acid ... more The mechanism of formation of the in-chain, unsaturated fatty acid metabolite, ⌬ 3-valproic acid (⌬ 3-VPA) by rat liver microsomes was examined. Microsomal rates of formation of ⌬ 3-VPA were below quantifiable limits in reactions catalyzed by control female rat liver microsomes, but were induced more than 20-fold following pretreatment with triacetyloleandomycin and pregnenolone-16␣-carbonitrile. Microsomal incubations conducted with 3-hydroxy-VPA or [2-2 H 1 ]VPA demonstrated that ⌬ 3-VPA did not arise by dehydration of preformed alcohol nor was it reversibly isomerized to ⌬ 2-VPA. CYP3A1 expression was optimized in the baculovirus expression vector system, and infected insect cell membranes which were supplemented with P450 reductase catalyzed formation of 3-OH-, 4-OH-, 5-OH-, ⌬ 3-, and ⌬ 4-VPA in ratios of 160:35: 6:3:1. Intramolecular deuterium isotope effects on metabolite formation, determined with cDNA-expressed CYP3A1 and either [3,3-2 H 2 ]VPA or [4,4-2 H 2 ]VPA, yielded k H /k D values for ⌬ 3-VPA of 2.00 ؎ 0.06 and 2.36 ؎ 0.08, respectively. These values were significantly lower than the isotope effects observed in the same incubations for 3-OH-VPA formation from 3,3-D 2-VPA (k H /k D ؍ 6.04 ؎ 0.08), or for 4-OH-and ⌬ 4-VPA formation from 4,4-D 2-VPA (k H /k D > 5). Collectively, these data demonstrate the existence of a microsomal P450-dependent in-chain fatty acid desaturase system distinct from the well-documented cytochrome b 5-linked CoA desaturases and suggest further that CYP3A1-dependent formation of ⌬ 3-VPA arises via nonselective, initial hydrogen atom abstraction from either the C-3 or the C-4 position.
The relevance of this to any discipline and ours in particular is that it offers an opportunity t... more The relevance of this to any discipline and ours in particular is that it offers an opportunity to look for patterns in our own scientific collaborations. We need ways to characterize what work is being carried out in an area, the impact it's having, who's being influenced by it, who are the sources of expertise on a particular topic, where the hubs are, and so on. These broad questions of scientific-knowledge management and of the structure and nature of scientific domains are also amenable to our methods and technologies. A prime example ...
We here describe research into the formal specification and implementation of resource-bounded ag... more We here describe research into the formal specification and implementation of resource-bounded agents. In particular, we provide an overview of our work on incorporating resource limitations into executable agent specifications. In addition, we outline future directions, highlighting both their promise and their problems.
There is increasing use of combinations of modal logics in both foundational and applied research... more There is increasing use of combinations of modal logics in both foundational and applied research areas. This article provides an introduction to both the principles of such combinations and to the variety of techniques that have been developed for them. In addition, the article outlines many key research problems yet to be tackled within this callenging area of work.
Proteome maps of hepatopancreas (midgut gland) and ovarian tissues of the crustacean, Cancer pagu... more Proteome maps of hepatopancreas (midgut gland) and ovarian tissues of the crustacean, Cancer pagurus (Decapoda; edible crab) have been produced by 2D-PAGE and identification of proteins, following trypsin proteolysis, by electrospray MS/MS and database searching. Owing to the lack of sequence information on proteins and fully sequenced genomes amongst the decapod crustaceans and given the evolutionary distance to the nearest full genome database (Daphnia), it was necessary to adopt a non-conventional identification approach. Thus, a strategy was developed for effective identification of decapod proteins by sequence similarity, homologybased cross-species database searching, using various algorithms and a combination of NCBI Crustacea and Arthropoda databases, together with the Arthropoda PartiGene database (Blaxter, University of Edinburgh). In both hepatopancreas and ovary tissues, the largest group of proteins identified were a variety of enzymes, followed by a smaller number of storage/ transport proteins [including vitellogenin (yolk protein), several subunits of hemocyanin, cryptocyanin, ferritin and calreticulin], with fewer structural proteins (actin, tubulin) and heatshock proteins, in addition to a number of proteins of miscellaneous functions. Such protein identifications allow the development of tools, such as antibodies and RNA/DNA probes, to investigate the functions of the proteins in specific tissues during development.
In this talk I shall try and argue that appropriate techniques based on formalisms based on modal... more In this talk I shall try and argue that appropriate techniques based on formalisms based on modal logic provide a powerful tool for the specification and verification of multiagent systems. The talk will consist of two parts. In the first part the idea of specifications of multiagent systems by means of temporal epistemic logic will be put forward and the main results presented. In the second, attention will be given to the problem of verifying that a multiagent system follows said specifications; the main techniques for verification by model checking (via OBDD and SAT) will be introduced and demonstrated.
This paper describes first steps towards the formal specification and verification of Distributed... more This paper describes first steps towards the formal specification and verification of Distributed Artificial Intelligence (DAI) systems, through the use of temporal belief logics. The paper first describes Concurrent METATEM, a programming language for DAI, and then develops a logic that may be used to reason about Concurrent METATEM systems. The utility of this logic for specifying and verifying Concurrent METATEM systems is demonstrated through a number of examples. The paper concludes with a brief discussion of the wider implications of the work, and in particular on the use of similar logics for reasoning about DAI systems in general.
Copyright is held by the author/owner(s). Submission for Review Article ACM Communications. these... more Copyright is held by the author/owner(s). Submission for Review Article ACM Communications. these questions, we will first recap autonomous systems (in Section 2) and formal verification (in Section 3).
The purpose of this paper is twofold: (i) to illustrate and re-emphasize the use of CONCURRENT ME... more The purpose of this paper is twofold: (i) to illustrate and re-emphasize the use of CONCURRENT METATEM, a programming language based on executable temporal logic, as a viable framework in which to develop Distributed Artificial Intelligence (DAI) applications; and (ii) to present a specific example of a cooperative protocol, give an outline of its implementation in CONCURRENT METATEM, and show how properties of the protocol can be established through formal proof.
METATEM is a simple programming language based on the direct execution of temporal logic statemen... more METATEM is a simple programming language based on the direct execution of temporal logic statements. It was introduced through a number of papers [35,2,3] culminating in a book collecting together work on the basic temporal language [5]. However, since that time, there has been a programme of research, carried out over a number of years, extending, adapting and applying the basic approach. In particular, much of the research has concerned the development of descendents of METATEM for describing and implementing complex multi-agent systems. Thus, while there are a number of other approaches to executing temporal statements [32,18], we will concentrate on this one particular approach and will describe the developments over the last 15 years. The structure of this article mirrors the research developments in that the path through these developments is not linear. The diagram below gives a pictorial explanation of the sections that follow.
The 1998 Workshop of the UK Special Interest Group on Multi-Agent Systems was held in Manchester ... more The 1998 Workshop of the UK Special Interest Group on Multi-Agent Systems was held in Manchester in December, chaired and organised by Michael Fisher of Manchester Metropolitan University, continuing the series of focussed and constructive meetings in this field. After two very successful workshops on the Foundations of Multi-Agent Systems at the University of Warwick in 1996 (Luck, 1997; Doran et al., 1997; d'Inverno et al., 1997; Fisher et al., 1997) and 1997 (Luck et al., 1998; Aylett et al., 1998; Binmore et al., 1998), the ...
Interest in agent-oriented technology continues to grow, both at a theoretical level and a practi... more Interest in agent-oriented technology continues to grow, both at a theoretical level and a practical level, with the UK maintaining a strong representation in the area in both academic institutions and commercial organisations. In December 1997, the Second UK Workshop on Foundations of Multi-Agent Systems (FoMAS'97), held at the University of Warwick, built on the success of FoMAS'96 a year earlier (Luck, 1997; Doran et al., 1997; d'Inverno et al., 1997; Fisher et al., 1997) in seeking to provide a forum for academics and industrialists within the UK to present and discuss current directions in research and applications development. Supported by the EPSRC and HP Labs, FoMAS'97 was expanded to two days with invited presentations from Professor Ken Binmore of UCL, and Professor Cristiano Castelfranchi of the Italian National Research Council's Institute of Psychology, in addition to paper presentations and panel discussions. The aim again was to provide an opportunity for promoting and supporting activity in the research and development of multi-agent systems. The two panels, which were concerned with Agent Systems and Applications, and Agent Rationality, comprising leading academics and industrialists, each generated stimulating and sometimes heated debates. Reports on these panel discussions follow separately. In this report, we summarise the other contributions to the workshop through paper presentations and invited talks, which cover a wide range of relevant topics. The structure of the report reflects the organisation of the workshop. The workshop began with an invited talk from Castelfranchi, who focussed on foundations of multi-agent systems with a social and psychological perspective. His starting point is the new AI of the '90s in which an important stream is artificial social intelligence. The main claims of the talk can be summarised as follows. First, the real foundation of all sociality (cooperation, competition, groups, organisation, etc.) is individual social action and mind. One cannot reduce or connect action at the collective level to action at the individual level unless one passes through the social character of individual action. Second, important levels of coordination and cooperation necessarily require minds and cognitive agents (beliefs, desires, intentions, etc.). However, cognition, communication and agreement are not enough for modelling and implementing cooperation: emergent pre-cognitive structures and constraints should be formalised, and emergent forms of cooperation are also needed among planning and deliberative agents. The final claim was that we are going towards a synthetic paradigm in AI and Cognitive Science, reconciling situatedness and plans, reactivity and mental representations, cognition, emergence and self-organisation. £ This report summarises the paper presentations at the Second UK Workshop on Foundations of Multi-Agent Systems.
As computer scientists, our goals are motivated by the desire to improve computer systems in some... more As computer scientists, our goals are motivated by the desire to improve computer systems in some way: making them easier to design and implement, more robust and less prone to error, easier to use, faster, cheaper, and so on. In the field of multi-agent systems, our goal is to build systems capable of flexible autonomous decision making, with societies of such systems cooperating with one-another. There is a lot of formal theory in the area but it is often not obvious what such theories should represent and what role the theory is ...
The mechanism of formation of the in-chain, unsaturated fatty acid metabolite, ⌬ 3-valproic acid ... more The mechanism of formation of the in-chain, unsaturated fatty acid metabolite, ⌬ 3-valproic acid (⌬ 3-VPA) by rat liver microsomes was examined. Microsomal rates of formation of ⌬ 3-VPA were below quantifiable limits in reactions catalyzed by control female rat liver microsomes, but were induced more than 20-fold following pretreatment with triacetyloleandomycin and pregnenolone-16␣-carbonitrile. Microsomal incubations conducted with 3-hydroxy-VPA or [2-2 H 1 ]VPA demonstrated that ⌬ 3-VPA did not arise by dehydration of preformed alcohol nor was it reversibly isomerized to ⌬ 2-VPA. CYP3A1 expression was optimized in the baculovirus expression vector system, and infected insect cell membranes which were supplemented with P450 reductase catalyzed formation of 3-OH-, 4-OH-, 5-OH-, ⌬ 3-, and ⌬ 4-VPA in ratios of 160:35: 6:3:1. Intramolecular deuterium isotope effects on metabolite formation, determined with cDNA-expressed CYP3A1 and either [3,3-2 H 2 ]VPA or [4,4-2 H 2 ]VPA, yielded k H /k D values for ⌬ 3-VPA of 2.00 ؎ 0.06 and 2.36 ؎ 0.08, respectively. These values were significantly lower than the isotope effects observed in the same incubations for 3-OH-VPA formation from 3,3-D 2-VPA (k H /k D ؍ 6.04 ؎ 0.08), or for 4-OH-and ⌬ 4-VPA formation from 4,4-D 2-VPA (k H /k D > 5). Collectively, these data demonstrate the existence of a microsomal P450-dependent in-chain fatty acid desaturase system distinct from the well-documented cytochrome b 5-linked CoA desaturases and suggest further that CYP3A1-dependent formation of ⌬ 3-VPA arises via nonselective, initial hydrogen atom abstraction from either the C-3 or the C-4 position.
The relevance of this to any discipline and ours in particular is that it offers an opportunity t... more The relevance of this to any discipline and ours in particular is that it offers an opportunity to look for patterns in our own scientific collaborations. We need ways to characterize what work is being carried out in an area, the impact it's having, who's being influenced by it, who are the sources of expertise on a particular topic, where the hubs are, and so on. These broad questions of scientific-knowledge management and of the structure and nature of scientific domains are also amenable to our methods and technologies. A prime example ...
We here describe research into the formal specification and implementation of resource-bounded ag... more We here describe research into the formal specification and implementation of resource-bounded agents. In particular, we provide an overview of our work on incorporating resource limitations into executable agent specifications. In addition, we outline future directions, highlighting both their promise and their problems.
There is increasing use of combinations of modal logics in both foundational and applied research... more There is increasing use of combinations of modal logics in both foundational and applied research areas. This article provides an introduction to both the principles of such combinations and to the variety of techniques that have been developed for them. In addition, the article outlines many key research problems yet to be tackled within this callenging area of work.
Proteome maps of hepatopancreas (midgut gland) and ovarian tissues of the crustacean, Cancer pagu... more Proteome maps of hepatopancreas (midgut gland) and ovarian tissues of the crustacean, Cancer pagurus (Decapoda; edible crab) have been produced by 2D-PAGE and identification of proteins, following trypsin proteolysis, by electrospray MS/MS and database searching. Owing to the lack of sequence information on proteins and fully sequenced genomes amongst the decapod crustaceans and given the evolutionary distance to the nearest full genome database (Daphnia), it was necessary to adopt a non-conventional identification approach. Thus, a strategy was developed for effective identification of decapod proteins by sequence similarity, homologybased cross-species database searching, using various algorithms and a combination of NCBI Crustacea and Arthropoda databases, together with the Arthropoda PartiGene database (Blaxter, University of Edinburgh). In both hepatopancreas and ovary tissues, the largest group of proteins identified were a variety of enzymes, followed by a smaller number of storage/ transport proteins [including vitellogenin (yolk protein), several subunits of hemocyanin, cryptocyanin, ferritin and calreticulin], with fewer structural proteins (actin, tubulin) and heatshock proteins, in addition to a number of proteins of miscellaneous functions. Such protein identifications allow the development of tools, such as antibodies and RNA/DNA probes, to investigate the functions of the proteins in specific tissues during development.
Uploads
Papers by Michael Fisher