Computer Science Department Faculty Publication Series

Permanent URI for this collection

Browse

Recent Submissions

  • Publication
    A CASE-BASED APPROACH TO MODELING LEGAL EXPERTISE
    (1988) ASHLEY, KD; Rissland, EL
  • Publication
    Case-based reasoning integrations
    (2002-01-01) Marling, C; Sqalli, M; Rissland, E; Munoz-Avila, H; Aha, D
  • Publication
    CABARET - RULE INTERPRETATION IN A HYBRID ARCHITECTURE
    (1991) Rissland, EL; SKALAK, DB
  • Publication
    The synergistic application of CBR to IR
    (1996) Rissland, EL; Daniels, JJ
  • Publication
    AI and Law: A fruitful synergy
    (2003-01-01) Rissland, EL; Ashley, KD; Loui, RP
  • Publication
    Law, learning and representation
    (2003-01-01) Ashley, KD; Rissland, EL
  • Publication
    AI and similarity
    (2006-01-01) Rissland, EL
  • Publication
    Integrations with case-based reasoning
    (2005-01-01) Marling, C; Rissland, E; Aamodt, A
  • Publication
    Case-based reasoning and law
    (2005-01-01) Rissland, EL; Ashley, KD; Branting, LK
  • Publication
    Autonomous Shaping: Knowledge Transfer in Reinforcement Learning
    (2006-01-01) Konidaris, George
    We introduce the use of learned shaping rewards in reinforcement learning tasks, where an agent uses prior experience on a sequence of tasks to learn a portable predictor that estimates intermediate rewards, resulting in accelerated learning in later tasks that are related but distinct. Such agents can be trained on a sequence of relatively easy tasks in order to develop a more informative measure of reward that can be transferred to improve performance on more difficult tasks without requiring a hand coded shaping function. We use a rod positioning task to show that this significantly improves performance even after a very brief training period.
  • Publication
    Privacy Vulnerabilities in Encrypted HTTP Streams
    (2005-01-01) Bissias, George Dean
    Encrypting traffic does not prevent an attacker from per- forming some types of traffic analysis. We present a straightforward traf- fic analysis attack against encrypted HTTP streams that is surprisingly effective in identifying the source of the traffic. An attacker starts by creating a profile of the statistical characteristics of web requests from interesting sites, including distributions of packet sizes and inter-arrival times. Later, candidate encrypted streams are compared against these profiles. In our evaluations using real traffic, we find that many web sites are subject to this attack. With a training period of 24 hours and a 1 hour delay afterwards, the attack achieves only 23% accuracy. However, an attacker can easily pre-determine which of trained sites are easily identifiable. Accordingly, against 25 such sites, the attack achieves 40% accuracy; with three guesses, the attack achieves 100% accuracy for our data. Longer delays after training decrease accuracy, but not substan- tially. We also propose some countermeasures and improvements to our current method. Previous work analyzed SSL traffic to a proxy, taking advantage of a known flaw in SSL that reveals the length of each web ob- ject. In contrast, we exploit the statistical characteristics of web streams that are encrypted as a single flow, which is the case with WEP/WPA, IPsec, and SSH tunnels.
  • Publication
    Exterminator: Automatically Correcting Memory Errors with High Probability
    (2007-06-01) Novark, Gene
    Programs written in C and C++ are susceptible to memory errors, including buffer overflows and dangling pointers. These errors, which can lead to crashes, erroneous execution, and security vulnerabilities, are notoriously costly to repair. Tracking down their location in the source code is difficult, even when the full memory state of the program is available. Once the errors are finally found, fixing them remains challenging: even for critical security-sensitive bugs, the average time between initial reports and the issuance of a patch is nearly one month. We present Exterminator, a system that automatically corrects heap-based memory errors without programmer intervention. Exterminator exploits randomization to pinpoint errors with high precision. From this information, Exterminator derives runtime patches that fix these errors both in current and subsequent executions. In addition, Exterminator enables collaborative bug correction by merging patches generated by multiple users. We present analytical and empirical results that demonstrate Exterminator’s effectiveness at detecting and correcting both injected and real faults.
  • Publication
    Black Swans, Gray Cygnets and Other Rare Birds
    (2009-01-01) Rissland, EL
    Surprising, exceptional cases — so-called black swans — can provoke extraordinary change in the way we do things or conceptualize the world. While it is not unreasonable to be surprised by a black swan, to be surprised by subsequent cases that are similar enough that they might cause the same sort of upheaval is unforgivable. The problem is how to reason about these almost novel, not totally unforeseen, subsequent cases that I call gray cygnets.
  • Publication
    Unsupervised Joint Alignment of Complex Images
    (2007-01-01) Huang, Gary B.
    Many recognition algorithms depend on careful positioning of an object into a canonical pose, so the position of features relative to a fixed coordinate system can be examined. Currently, this positioning is done either manually or by training a class-specialized learning algorithm with samples of the class that have been hand-labeled with parts or poses. In this paper, we describe a novel method to achieve this positioning using poorly aligned examples of a class with no additional labeling. Given a set of unaligned examplars of a class, such as faces, we automatically build an alignment mechanism, without any additional labeling of parts or poses in the data set. Using this alignment mechanism, new members of the class, such as faces resulting from a face detector, can be precisely aligned for the recognition process. Our alignment method improves performance on a face recognition task, both over unaligned images and over images aligned with a face alignment algorithm specifically developed for and trained on hand-labeled face images. We also demonstrate its use on an entirely different class of objects (cars), again without providing any information about parts or pose to the learning algorithm.
  • Publication
    Ultra-Low Power Data Storage for Sensor Networks
    (2006-01-01) Mathur, Gaurav
    Local storage is required in many sensor network applications, both for archival of detailed event information, as well as to overcome sensor platform memory constraints. While extensive measurement studies have been performed to highlight the trade-off between computation and communication in sensor networks, the role of storage has received little attention. The storage subsystems on currently available sensor platforms have not exploited technology trends, and consequently the energy cost of storage on these platforms is as high as that of communication. Current flash memories, however, offer a low-priced, high-capacity and extremely energy-efficient storage solution. In this paper, we perform a comprehensive evaluation of the active and sleep-mode energy consumption of available flash-based storage options for sensor platforms. Our results demonstrate more than a 100-fold decrease in per-byte energy consumption for surface-mount parallel NAND flash in comparison with the MicaZ on-board serial flash. In addition, this dramatically reduces storage energy costs relative to communication, introducing a new dimension in traditional computation vs communication trade-offs. Our results have significant ramifications on the design of sensor platforms as well as on the energy consumption of sensing applications. We quantify the potential energy gains for two commonly used sensor network services: communication and in-network data aggregation. Our measurements show significant improvements in each service: 50-fold and up to 10-fold reductions in energy for communication and data aggregation respectively.
  • Publication
    Automated fault tree analysis for engineering design optimization
    (2008-01-01) Shao, TF; Lin, ZF; Krishnamurty, S; Grosse, IR; Osterweil, LJ
    This paper presents an automated fault tree analysis for engineering design optimization process. Specifically, a novel approach is presented in which Little-JIL, a process programming language, is applied to create a process model of engineering optimization. The process model uses a graphical language in the form of easy-to-understand block diagrams for defining processes that coordinate the activities of autonomous agents and their use of resources during the performance of a task. The use of Little-JIL facilitates agent coordination in the design optimization process and helps to model the order of and the communications between units of sub-processes. The resulting process model is easy to debug and is rigorous for simulation and formal reasoning in engineering design optimization. Furthermore, it enables the development of a clear and precise design optimization process model at different levels of granularity as perhaps preferred by the user. Moreover, since the process model allows for generation of fault trees automatically, it can be expected to be less errorprone than manually generated ones. A case study is shown to demonstrate the effectiveness and efficiency of the automated fault tree approach to design optimization and its usefulness in engineering decision making and in improving reliability of engineering design process.