Browsing by Subject "Explanation"
- Results Per Page
- Sort Options
Item Open Access Answering and Explaining SQL Queries Privately(2022) Tao, YuchaoData privacy has been receiving an increasing amount of attention in recent years. While large-scale personal information is collected for scientific research and commercial products, a privacy breach is not acceptable as a trade-off. In the last decade, differential privacy has become a gold standard to protect data privacy and has been applied in many organizations. Past work focused on building a differentially private SQL query answering system as a building block for wider applications. However, answering counting queries with joins under differential privacy appears as a challenge. The join operator allows any user to have an unbounded impact on the query result, which impedes hiding the existence of a single user by differential privacy. On the other hand, the introduction of differential privacy to the query answering also prevents the users from understanding the query results correctly, since she needs to distinguish the effect of differential privacy from the contribution of data.
In this thesis, we study two problems about answering and explaining SQL queries privately. First, we present efficient algorithms to compute local sensitivities of counting queries with joins, which is an important premise for answering these queries under differential privacy. We track the sensitivity contributed by each tuple, based on which we propose a truncation mechanism that answers counting queries with joins privately with high utility. Second, we propose a formal framework DPXPlain, a three-phase framework that allows users to get explanations for group-by COUNT/SUM/AVG query results while preserving DP. We utilize confidence intervals to help users understand the uncertainty in the query results introduced by differential privacy, and further provide top-k explanations under differential privacy to explain the contribution of data to the results.
Item Open Access Understanding Cognition(2015) Steenbergen, Gordon J.Cognitive neuroscience is an interdisciplinary enterprise aimed at explaining cognition and behavior. It appears to be succeeding. What accounts for this apparent explanatory success? According to one prominent philosophical thesis, cognitive neuroscience explains by discovering and describing mechanisms. This "mechanist thesis" is open to at least two interpretations: a strong metaphysical thesis that Carl Craver and David Kaplan defend, and a weaker methodological thesis that William Bechtel defends. I argue that the metaphysical thesis is false and that the methodological thesis is too weak to account for the explanatory promise of cognitive neuroscience. My argument draws support from a representative example of research in this field, namely, the neuroscience of decision-making. The example shows that cognitive neuroscience explains in a variety of ways and that the discovery of mechanisms functions primarily as a way of marshaling evidence in support of the models of cognition that are its principle unit of explanatory significance.
The inadequacy of the mechanist program is symptomatic of an implausible but prominent view of scientific understanding. On this view, scientific understanding consists in an accurate and complete description of certain "objective" explanatory relations, that is, relations that hold independently of facts about human psychology. I trace this view to Carl Hempel's logical empiricist reconceptualization of scientific understanding, which then gets extended in Wesley Salmon's causal-mechanistic approach. I argue that the twin objectivist ideals of accuracy and completeness are neither ends we actually value nor ends we ought to value where scientific understanding is concerned.
The case against objectivism motivates psychologism about understanding, the view that understanding depends on human psychology. I propose and defend a normative psychologistic framework for investigating the nature of understanding in the mind sciences along three empirically-informed dimensions: 1) What are the ends of understanding? 2) What is the nature of the cognitive strategy that we deploy to achieve those ends; and 3) Under what conditions is our deployment of this strategy effective toward achieving those ends? To articulate and defend this view, I build on the work of Elliot Sober to develop a taxonomy of psychologisms about understanding. Epistemological psychologism, a species of naturalism, is the view that justifying claims about understanding requires appealing to what scientists actually do when they seek understanding. Metaphysical psychologism is the view that the truth-makers for claims about understanding include facts about human psychology. I defend both views against objections.