Sean Stromsten - Academia.edu (original) (raw)
Papers by Sean Stromsten
Lecture Notes in Computer Science, 2000
We examine the role of simplicity in directing the induction of context-free grammars from sample... more We examine the role of simplicity in directing the induction of context-free grammars from sample sentences. We present a rational reconstruction of Wolff's SNPR-the Grids system-which incorporates a bias toward grammars that minimize description length. The algorithm alternates between merging existing nonterminal symbols and creating new symbols, using a beam search to move from complex to simpler grammars. Experiments suggest that this approach can induce accurate grammars and that it scales reasonably to more difficult domains.
2009 IEEE Conference on Technologies for Homeland Security, 2009
This paper describes the theoretical basis and practical implementation of PARSEC, a knowledge-ba... more This paper describes the theoretical basis and practical implementation of PARSEC, a knowledge-based system that uses probabilistic case based reasoning. PARSEC is a major component in the PANDA surveillance system, developed under DARPA leadership to support maritime situation awareness monitoring on a global scale. PANDA detects unusual vessel motions (deviations) based on learned normalcy models and then flags those particular
There is growing interest in automating the detection of interesting new developments in science ... more There is growing interest in automating the detection of interesting new developments in science and technology. BAE Systems is pursuing ARBITER (Abductive Reasoning Based on Indicators and Topics of EmeRgence), a multi-disciplinary study and development effort to analyze full- text and metadata for indicators of emergent technologies and scientific fields. To define these indicators, our team has applied the primary insights of actant network theory developed within the disciplines of Science and Technology Studies and the history of technology and science to create a pragmatic theory of technoscientific emergence. Specifically, this practical theory articulates emergence in terms of the robustness of actant networks. This applied actant-network theory currently guides our definition of indicators and indicator patterns for the ARBITER system, and represents a novel contribution to the discussion of emergent technologies and fields. Several elements of our theory were validated with 15 case studies and 25 example technologies.
Neural Computation, 2007
We propose a new method, parametric embedding (PE), that embeds objects with the class structure ... more We propose a new method, parametric embedding (PE), that embeds objects with the class structure into a low-dimensional visualization space. PE takes as input a set of class conditional probabilities for given data points and tries to preserve the structure in an embedding space by minimizing a sum of Kullback-Leibler divergences, under the assumption that samples are generated by a gaussian mixture with equal covariances in the embedding space. PE has many potential uses depending on the source of the input data, providing insight into the classifier's behavior in supervised, semisupervised, and unsupervised settings. The PE algorithm has a computational advantage over conventional embedding methods based on pairwise object relations since its complexity scales with the product of the number of objects and the number of classes. We demonstrate PE by visualizing supervised categorization of Web pages, semisupervised categorization of digits, and the relations of words and latent...
Let the `skeleton'of T be the subtree consisting of all paths from the label... more Let the `skeleton'of T be the subtree consisting of all paths from the labeled leaves to the root. Since the mutation process is conservative, the classification of any node N<SUB>i according to TBB is the most likely value at the node where the path from N<SUB>i meets the skeleton. Let N<SUB>j be any labeled node, N<SUB>L be the set of all labeled nodes, and N L j be the set of all labeled nodes except N<SUB>j.
Advances in Neural Information Processing Systems, 2004
We describe a nonparametric Bayesian approach to generalizing from few labeled examples, guided b... more We describe a nonparametric Bayesian approach to generalizing from few labeled examples, guided by a larger set of unlabeled objects and the assumption of a latent tree-structure to the domain. The tree (or a distribution over trees) may be inferred using the unlabeled data. A prior over concepts generated by a mutation process on the inferred tree (s) allows efficient computation of the optimal Bayesian classification function from the labeled examples. We test our approach on eight real-world datasets.
Recent advances in normative frameworks for causal reasoning have inspired psychologists to model... more Recent advances in normative frameworks for causal reasoning have inspired psychologists to model human causal inference using these normative frameworks. These models are, however, vulnerable to the same criticisms as connectionist and other probabilistic models. A proposed response to these criticisms is the notion of first-order causal rules, which are a simple generalization of dependencies in causal Bayes nets, adding universally quantified logical variables. Psychology adopts normative theories
Lecture Notes in Computer Science, 2000
We examine the role of simplicity in directing the induction of context-free grammars from sample... more We examine the role of simplicity in directing the induction of context-free grammars from sample sentences. We present a rational reconstruction of Wolff's SNPR-the Grids system-which incorporates a bias toward grammars that minimize description length. The algorithm alternates between merging existing nonterminal symbols and creating new symbols, using a beam search to move from complex to simpler grammars. Experiments suggest that this approach can induce accurate grammars and that it scales reasonably to more difficult domains.
2009 IEEE Conference on Technologies for Homeland Security, 2009
This paper describes the theoretical basis and practical implementation of PARSEC, a knowledge-ba... more This paper describes the theoretical basis and practical implementation of PARSEC, a knowledge-based system that uses probabilistic case based reasoning. PARSEC is a major component in the PANDA surveillance system, developed under DARPA leadership to support maritime situation awareness monitoring on a global scale. PANDA detects unusual vessel motions (deviations) based on learned normalcy models and then flags those particular
There is growing interest in automating the detection of interesting new developments in science ... more There is growing interest in automating the detection of interesting new developments in science and technology. BAE Systems is pursuing ARBITER (Abductive Reasoning Based on Indicators and Topics of EmeRgence), a multi-disciplinary study and development effort to analyze full- text and metadata for indicators of emergent technologies and scientific fields. To define these indicators, our team has applied the primary insights of actant network theory developed within the disciplines of Science and Technology Studies and the history of technology and science to create a pragmatic theory of technoscientific emergence. Specifically, this practical theory articulates emergence in terms of the robustness of actant networks. This applied actant-network theory currently guides our definition of indicators and indicator patterns for the ARBITER system, and represents a novel contribution to the discussion of emergent technologies and fields. Several elements of our theory were validated with 15 case studies and 25 example technologies.
Neural Computation, 2007
We propose a new method, parametric embedding (PE), that embeds objects with the class structure ... more We propose a new method, parametric embedding (PE), that embeds objects with the class structure into a low-dimensional visualization space. PE takes as input a set of class conditional probabilities for given data points and tries to preserve the structure in an embedding space by minimizing a sum of Kullback-Leibler divergences, under the assumption that samples are generated by a gaussian mixture with equal covariances in the embedding space. PE has many potential uses depending on the source of the input data, providing insight into the classifier's behavior in supervised, semisupervised, and unsupervised settings. The PE algorithm has a computational advantage over conventional embedding methods based on pairwise object relations since its complexity scales with the product of the number of objects and the number of classes. We demonstrate PE by visualizing supervised categorization of Web pages, semisupervised categorization of digits, and the relations of words and latent...
Let the `skeleton'of T be the subtree consisting of all paths from the label... more Let the `skeleton'of T be the subtree consisting of all paths from the labeled leaves to the root. Since the mutation process is conservative, the classification of any node N<SUB>i according to TBB is the most likely value at the node where the path from N<SUB>i meets the skeleton. Let N<SUB>j be any labeled node, N<SUB>L be the set of all labeled nodes, and N L j be the set of all labeled nodes except N<SUB>j.
Advances in Neural Information Processing Systems, 2004
We describe a nonparametric Bayesian approach to generalizing from few labeled examples, guided b... more We describe a nonparametric Bayesian approach to generalizing from few labeled examples, guided by a larger set of unlabeled objects and the assumption of a latent tree-structure to the domain. The tree (or a distribution over trees) may be inferred using the unlabeled data. A prior over concepts generated by a mutation process on the inferred tree (s) allows efficient computation of the optimal Bayesian classification function from the labeled examples. We test our approach on eight real-world datasets.
Recent advances in normative frameworks for causal reasoning have inspired psychologists to model... more Recent advances in normative frameworks for causal reasoning have inspired psychologists to model human causal inference using these normative frameworks. These models are, however, vulnerable to the same criticisms as connectionist and other probabilistic models. A proposed response to these criticisms is the notion of first-order causal rules, which are a simple generalization of dependencies in causal Bayes nets, adding universally quantified logical variables. Psychology adopts normative theories