We describe a relational learning by observation framework that automatically creates cognitive a... more We describe a relational learning by observation framework that automatically creates cognitive agent programs that model expert task performance in complex dynamic domains. Our framework uses observed behavior and goal annotations of an expert as the primary input, interprets them in the context of background knowledge, and returns an agent program that behaves similar to the expert. We map the problem of creating an agent program on to multiple learning problems that can be represented in a “supervised concept learning” setting. The acquired procedural knowledge is partitioned into a hierarchy of goals and represented with first order rules. Using an inductive logic programming (ILP) learning component allows our framework to naturally combine structured behavior observations, parametric and hierarchical goal annotations, and complex background knowledge. To deal with the large domains we consider, we have developed an efficient mechanism for storing and retrieving structured beha...
Proceedings of the 23rd international …, Jan 1, 2006
Knowledge-based planning methods offer benefits over classical techniques, but they are time cons... more Knowledge-based planning methods offer benefits over classical techniques, but they are time consuming and costly to construct. There has been research on learning plan knowledge from search, but this can take substantial computer time and may even fail to find solutions on complex tasks. Here we describe another approach that observes sequences of operators taken from expert solutions to problems and learns hierarchical task networks from them. The method has similarities to previous algorithms for explanation-based learning, but differs in its ability to acquire hierarchical structures and in the generality of learned conditions. These increase the method's capability to transfer learned knowledge to other problems and supports the acquisition of recursive procedures. After presenting the learning algorithm, we report experiments that compare its abilities to other techniques on two planning domains. In closing, we review related work and directions for future research.
We describe a relational learning by observation framework that automatically creates cognitive a... more We describe a relational learning by observation framework that automatically creates cognitive agent programs that model expert task performance in complex dynamic domains. Our framework uses observed behavior and goal annotations of an expert as the primary input, interprets them in the context of background knowledge, and returns an agent program that behaves similar to the expert. We map the problem of creating an agent program on to multiple learning problems that can be represented in a "supervised concept learning" setting. The acquired procedural knowledge is partitioned into a hierarchy of goals and represented with first order rules. Using an inductive logic programming (ILP) learning component allows our framework to naturally combine structured behavior observations, parametric and hierarchical goal annotations, and complex background knowledge. To deal with the large domains we consider, we have developed an efficient mechanism for storing and retrieving structured behavior data. We have tested our approach using artificially created examples and behavior observation traces generated by AI agents. We evaluate the learned rules by comparing them to hand-coded rules.
In this paper, we present a principled approach to constructing believable game players that reli... more In this paper, we present a principled approach to constructing believable game players that relies on a cognitive architecture. The resulting agent is capable of playing the game Urban Combat in a plausible manner when faced with similar situations as its human counterparts. We discuss how architectural features like goal-directed but reactive execution and incremental learning can produce more believable synthetic characters.
Qualitative Reasoning: The Twelfth International …, Jan 1, 1998
We show that qualitative simulation algorithms can make better use of their input to deduce signi... more We show that qualitative simulation algorithms can make better use of their input to deduce significant amounts of information about the relative lengths of the time intervals in their output behavior predictions . Simple techniques employing concepts like symmetry and periodicity, and comparison of the circumstances during multiple traversals ofthe same interval can enable the reasoner to build a list of facts representing the deduced information about relative durations . These facts are used by a new filter, which eliminates proposed spurious behaviors leading to inconsistent duration data. Surviving behaviors are annotated with richer descriptions of the qualitative properties of system variables, in addition to the extracted relative duration information .
Developing computer game agents is often a lengthy and expensive undertaking. Detailed domain kno... more Developing computer game agents is often a lengthy and expensive undertaking. Detailed domain knowledge and decision-making procedures must be encoded into the agent to achieve realistic behavior. In this paper, we simplify this process by using the ICARUS cognitive architecture to construct game agents. The system acquires structured, high fidelity methods for agents that utilize a vocabulary of concepts familiar to game experts. We demonstrate our approach by first acquiring behaviors for football agents from video footage of college football games, and then applying the agents in a football simulator.
Proceedings of the Twenty-Third AAAI …, Jan 1, 2008
Transfer is the ability to employ knowledge acquired in one task to improve performance in anothe... more Transfer is the ability to employ knowledge acquired in one task to improve performance in another. We study transfer in the context of the ICARUS cognitive architecture, which supplies diverse capabilities for execution, inference, planning, and learning. We report on an extension to ICARUS called representation mapping that transfers structured skills and concepts between disparate tasks that may not even be expressed with the same symbol set. We show that representation mapping is naturally integrated into ICARUS' cognitive processing loop, resulting in a system that addresses a qualitatively new class of problems by considering the relevance of past experience to current goals.
With the enormous success of first-person perspective games, developers have realized the importa... more With the enormous success of first-person perspective games, developers have realized the importance of believable agents, partly due to increasing user expectations. Until recently, automated agents in these games have been either given too much information or restricted to limited behaviors. In this paper, we present an architectural response to this problem, using a cognitive architecture to construct an intelligent agent. The resulting agent is capable of playing one such game, Urban Combat, in more believable manner, when faced with similar situations as its human counterparts. We discuss how an architecture with embedded execution and learning capabilities can embody a more believable game player that can be an important, and enjoyable part of a game.
In this paper, we present an approach to transfer that involves analogical mapping of symbols acr... more In this paper, we present an approach to transfer that involves analogical mapping of symbols across different domains. We relate this mechanism to ICARUS, a theory of the human cognitive architecture. Our system can transfer skills across domains hypothesizing maps between representations, improving performance in novel domains. Unlike previous approaches to analogical transfer, our method uses an explanatory analysis that compares how well a new domain theory explains previous solutions under different mapping hypotheses. We present experimental evidence that the new mechanism improves transfer over ICARUS' basic learning processes. Moreover, we argue that the same features which distinguish ICARUS from other architectures support representation mapping in a natural way and operate synergistically with it. These features enable our analogy system to translate a map among concepts into a map between skills, and to support transfer even if two domains are only partially analogous. We also discuss our system's relation to other work on analogy and outline directions for future research.
Proceedings of the Symposium on Mixed- …, Jan 1, 2005
We describe a learning from diagrammatic behavior specifications approach, where the task-perform... more We describe a learning from diagrammatic behavior specifications approach, where the task-performance knowledge of a human expert is transferred to an agent program using abstract behavior scenarios that the expert and the agent program interactively specify. The diagrammatic interface serves as a communication medium between the expert and the agent program to share knowledge during behavior specification. A relational learning by observation component interprets these scenarios in the context of background knowledge and expert annotations to learn first-order rules that represent the task-performance knowledge for an improved agent program.
Annals of Mathematics and Artificial Intelligence, Jan 1, 2003
We present two new qualitative reasoning formalisms, and use them in the construction of a new ty... more We present two new qualitative reasoning formalisms, and use them in the construction of a new type of filtering mechanism for qualitative simulators. Our new sign algebra, SR1*, facilitates reasoning about relationships among the signs of collections of real numbers. The comparison calculus, built on top of SR1*, is a general framework that can be used to qualitatively compare the behaviors of two dynamic systems or two excerpts of the behavior of a single dynamic system at different situations. These tools enable us to improve the predictive performance of qualitative simulation algorithms. We show that qualitative simulators can make better use of their input to deduce significant amounts of qualitative information about the relative lengths of the time intervals in their output behavior predictions. Simple techniques employing concepts like symmetry, periodicity, and comparison of the circumstances during multiple traversals of the same region can be used to build a list of facts representing the deduced information about relative durations. The duration consistency filter eliminates spurious behaviors leading to inconsistent combinations of these facts. Surviving behaviors are annotated with richer qualitative descriptions. Used in conjunction with other spurious behavior elimination methods, this approach would increase the ability of qualitative simulators to handle more complex systems.
We describe a relational learning by observation framework that automatically creates cognitive a... more We describe a relational learning by observation framework that automatically creates cognitive agent programs that model expert task performance in complex dynamic domains. Our framework uses observed behavior and goal annotations of an expert as the primary input, interprets them in the context of background knowledge, and returns an agent program that behaves similar to the expert. We map the problem of creating an agent program on to multiple learning problems that can be represented in a “supervised concept learning” setting. The acquired procedural knowledge is partitioned into a hierarchy of goals and represented with first order rules. Using an inductive logic programming (ILP) learning component allows our framework to naturally combine structured behavior observations, parametric and hierarchical goal annotations, and complex background knowledge. To deal with the large domains we consider, we have developed an efficient mechanism for storing and retrieving structured beha...
Proceedings of the 23rd international …, Jan 1, 2006
Knowledge-based planning methods offer benefits over classical techniques, but they are time cons... more Knowledge-based planning methods offer benefits over classical techniques, but they are time consuming and costly to construct. There has been research on learning plan knowledge from search, but this can take substantial computer time and may even fail to find solutions on complex tasks. Here we describe another approach that observes sequences of operators taken from expert solutions to problems and learns hierarchical task networks from them. The method has similarities to previous algorithms for explanation-based learning, but differs in its ability to acquire hierarchical structures and in the generality of learned conditions. These increase the method's capability to transfer learned knowledge to other problems and supports the acquisition of recursive procedures. After presenting the learning algorithm, we report experiments that compare its abilities to other techniques on two planning domains. In closing, we review related work and directions for future research.
We describe a relational learning by observation framework that automatically creates cognitive a... more We describe a relational learning by observation framework that automatically creates cognitive agent programs that model expert task performance in complex dynamic domains. Our framework uses observed behavior and goal annotations of an expert as the primary input, interprets them in the context of background knowledge, and returns an agent program that behaves similar to the expert. We map the problem of creating an agent program on to multiple learning problems that can be represented in a "supervised concept learning" setting. The acquired procedural knowledge is partitioned into a hierarchy of goals and represented with first order rules. Using an inductive logic programming (ILP) learning component allows our framework to naturally combine structured behavior observations, parametric and hierarchical goal annotations, and complex background knowledge. To deal with the large domains we consider, we have developed an efficient mechanism for storing and retrieving structured behavior data. We have tested our approach using artificially created examples and behavior observation traces generated by AI agents. We evaluate the learned rules by comparing them to hand-coded rules.
In this paper, we present a principled approach to constructing believable game players that reli... more In this paper, we present a principled approach to constructing believable game players that relies on a cognitive architecture. The resulting agent is capable of playing the game Urban Combat in a plausible manner when faced with similar situations as its human counterparts. We discuss how architectural features like goal-directed but reactive execution and incremental learning can produce more believable synthetic characters.
Qualitative Reasoning: The Twelfth International …, Jan 1, 1998
We show that qualitative simulation algorithms can make better use of their input to deduce signi... more We show that qualitative simulation algorithms can make better use of their input to deduce significant amounts of information about the relative lengths of the time intervals in their output behavior predictions . Simple techniques employing concepts like symmetry and periodicity, and comparison of the circumstances during multiple traversals ofthe same interval can enable the reasoner to build a list of facts representing the deduced information about relative durations . These facts are used by a new filter, which eliminates proposed spurious behaviors leading to inconsistent duration data. Surviving behaviors are annotated with richer descriptions of the qualitative properties of system variables, in addition to the extracted relative duration information .
Developing computer game agents is often a lengthy and expensive undertaking. Detailed domain kno... more Developing computer game agents is often a lengthy and expensive undertaking. Detailed domain knowledge and decision-making procedures must be encoded into the agent to achieve realistic behavior. In this paper, we simplify this process by using the ICARUS cognitive architecture to construct game agents. The system acquires structured, high fidelity methods for agents that utilize a vocabulary of concepts familiar to game experts. We demonstrate our approach by first acquiring behaviors for football agents from video footage of college football games, and then applying the agents in a football simulator.
Proceedings of the Twenty-Third AAAI …, Jan 1, 2008
Transfer is the ability to employ knowledge acquired in one task to improve performance in anothe... more Transfer is the ability to employ knowledge acquired in one task to improve performance in another. We study transfer in the context of the ICARUS cognitive architecture, which supplies diverse capabilities for execution, inference, planning, and learning. We report on an extension to ICARUS called representation mapping that transfers structured skills and concepts between disparate tasks that may not even be expressed with the same symbol set. We show that representation mapping is naturally integrated into ICARUS' cognitive processing loop, resulting in a system that addresses a qualitatively new class of problems by considering the relevance of past experience to current goals.
With the enormous success of first-person perspective games, developers have realized the importa... more With the enormous success of first-person perspective games, developers have realized the importance of believable agents, partly due to increasing user expectations. Until recently, automated agents in these games have been either given too much information or restricted to limited behaviors. In this paper, we present an architectural response to this problem, using a cognitive architecture to construct an intelligent agent. The resulting agent is capable of playing one such game, Urban Combat, in more believable manner, when faced with similar situations as its human counterparts. We discuss how an architecture with embedded execution and learning capabilities can embody a more believable game player that can be an important, and enjoyable part of a game.
In this paper, we present an approach to transfer that involves analogical mapping of symbols acr... more In this paper, we present an approach to transfer that involves analogical mapping of symbols across different domains. We relate this mechanism to ICARUS, a theory of the human cognitive architecture. Our system can transfer skills across domains hypothesizing maps between representations, improving performance in novel domains. Unlike previous approaches to analogical transfer, our method uses an explanatory analysis that compares how well a new domain theory explains previous solutions under different mapping hypotheses. We present experimental evidence that the new mechanism improves transfer over ICARUS' basic learning processes. Moreover, we argue that the same features which distinguish ICARUS from other architectures support representation mapping in a natural way and operate synergistically with it. These features enable our analogy system to translate a map among concepts into a map between skills, and to support transfer even if two domains are only partially analogous. We also discuss our system's relation to other work on analogy and outline directions for future research.
Proceedings of the Symposium on Mixed- …, Jan 1, 2005
We describe a learning from diagrammatic behavior specifications approach, where the task-perform... more We describe a learning from diagrammatic behavior specifications approach, where the task-performance knowledge of a human expert is transferred to an agent program using abstract behavior scenarios that the expert and the agent program interactively specify. The diagrammatic interface serves as a communication medium between the expert and the agent program to share knowledge during behavior specification. A relational learning by observation component interprets these scenarios in the context of background knowledge and expert annotations to learn first-order rules that represent the task-performance knowledge for an improved agent program.
Annals of Mathematics and Artificial Intelligence, Jan 1, 2003
We present two new qualitative reasoning formalisms, and use them in the construction of a new ty... more We present two new qualitative reasoning formalisms, and use them in the construction of a new type of filtering mechanism for qualitative simulators. Our new sign algebra, SR1*, facilitates reasoning about relationships among the signs of collections of real numbers. The comparison calculus, built on top of SR1*, is a general framework that can be used to qualitatively compare the behaviors of two dynamic systems or two excerpts of the behavior of a single dynamic system at different situations. These tools enable us to improve the predictive performance of qualitative simulation algorithms. We show that qualitative simulators can make better use of their input to deduce significant amounts of qualitative information about the relative lengths of the time intervals in their output behavior predictions. Simple techniques employing concepts like symmetry, periodicity, and comparison of the circumstances during multiple traversals of the same region can be used to build a list of facts representing the deduced information about relative durations. The duration consistency filter eliminates spurious behaviors leading to inconsistent combinations of these facts. Surviving behaviors are annotated with richer qualitative descriptions. Used in conjunction with other spurious behavior elimination methods, this approach would increase the ability of qualitative simulators to handle more complex systems.
Uploads
Papers by Tolga Konik