Darren Scott Appling

An Ensemble Architecture for Learning Complex Problem-Solving Techniques from Demonstration

We present a novel ensemble architecture for learning problem-solving techniques from a very small number of expert solutions and demonstrate its effectiveness in a complex real-world domain. The key feature of our “Generalized Integrated Learning Architecture” (GILA) is a set of heterogeneous independent learning and reasoning (ILR) components, coordinated by a central meta-reasoning executive (MRE). The ILRs are weakly coupled in the sense that all coordination during learning and performance happens through the MRE. Each ILR learns independently from a small number of expert demonstrations of a complex task. During performance, each ILR proposes partial solutions to subproblems posed by the MRE, which are then selected from and pieced together by the MRE to produce a complete solution. The heterogeneity of the learner-reasoners allows both learning and problem solving to be more effective because their abilities and biases are complementary and synergistic. We describe the application of this novel learning and problem solving architecture to the domain of airspace management, where multiple requests for the use of airspaces need to be deconflicted, reconciled, and managed automatically. Formal evaluations show that our system performs as well as or better than humans after learning from the same training data. Furthermore, GILA outperforms any individual ILR run in isolation, thus demonstrating the power of the ensemble architecture for learning and problem solving.

Discriminative Models for Predicting Deception Strategies

Although a large body of work has previously investigated various cues predicting deceptive communications, especially as demonstrated through written and spoken language (e.g., [30]), little has been done to explore predicting kinds of de- ception. We present novel work to evaluate the use of textual cues to discriminate between deception strategies (such as exaggeration or falsification), concentrating on intention- ally untruthful statements meant to persuade in a social media context. We conduct human subjects experimenta- tion wherein subjects were engaged in a conversational task and then asked to label the kind(s) of deception they employed for each deceptive statement made. We then develop discriminative models to understand the difficulty between choosing between one and several strategies. We evaluate the models using precision and recall for strategy prediction among 4 deception strategies based on the most relevant psycholinguistic, structural, and data-driven cues. Our single strategy model results demonstrate as much as a 58% increase over baseline (random chance) accuracy and we also find that it is more difficult to predict certain kinds of de- ception than others.

Social Network Derived Credibility

The increasing use of social media results in users that must ascertain the truthfulness of information that they encounter from unknown sources using a variety of indicators (e.g., explicit ratings, profile information, etc.). Through human-subject experimentation with an online social network-style platform, we focus on the determination of credibility in ego-centric networks, where participants are able to observe salient social network properties, such as degree centrality and geodesic distance. Using manipulated social network graphs, we find that corroboration and degree centrality are most utilized by subjects as indicators of credibility. utilized by subjects as indicators of credibility. We discuss the implications of the use of social network structural properties, use principal components analysis to visualize the reduced dimensional feature space, and analyze how credibility changes per property according to the “Big 5” theory of personality.