Hans Chalupsky

Simplifying construction of complex workflows for non-expert users of the Southern California Earthquake Center Community Modeling Environment

Workflow systems often present the user with rich interfaces that express all the capabilities and complexities of the application programs and the computing environments that they support. However, non-expert users are better served with simple interfaces that abstract away system complexities and still enable them to construct and execute complex workflows. To explore this idea, we have created a set of tools and interfaces that simplify the construction of workflows. Implemented as part of the Community Modeling Environment developed by the Southern California Earthquake Center, these tools, are integrated into a comprehensive workflow system that supports both domain experts as well as non expert users.

Electric Elves: Agent Technology for Supporting Human Organizations

The operation of a human organization requires dozens of everyday tasks to ensure coherence in organizational activities, to monitor the status of such activities, to gather information relevant to the organization, to keep everyone in the organization informed, etc. Teams of software agents can aid humans in accomplishing these tasks, facilitating the organization’s coherent functioning and rapid response to crises, while reducing the burden on humans. Based on this vision, this paper reports on Electric Elves, a system that has been operational, 24/7, at our research institute since June 1, 2000. Tied to individual user workstations, fax machines, voice, mobile devices such as cell phones and palm pilots, Electric Elves has assisted us in routine tasks, such as rescheduling meetings, selecting presenters for research meetings, tracking people’s locations, organizing lunch meetings, etc. We discuss the underlying AI technologies that led to the success of Electric Elves, including technologies devoted to agenthuman interactions, agent coordination, accessing multiple heterogeneous information sources, dynamic assignment of organizational tasks, and deriving information about organization members. We also report the results of deploying Electric Elves in our own research organization.

Processing-in-memory technology for knowledge discovery algorithms

The goal of this work is to gain insight into whether processing-in-memory (PIM) technology can be used to accelerate the performance of link discovery algorithms, which represent an important class of emerging knowledge discovery techniques. PIM chips that integrate processor logic into memory devices offer a new opportunity for bridging the growing gap between processor and memory speeds, especially for applications with high memory-bandwidth requirements. As LD algorithms are data-intensive and highly parallel, involving read-only queries over large data sets, parallel computing power extremely close (physically) to the data has the potential of providing dramatic computing speedups. For this reason, we evaluated the mapping of LD algorithms to a processing-in-memory (PIM) workstation-class architecture, the DIVA/Godiva hardware testbeds developed by USC/ISI. Accounting for differences in clock speed and data scaling, our analysis shows a performance gain on a single PIM, with the potential for greater improvement when multiple PIMs are used. Measured speedups of 8x are shown on two additional bandwidth benchmarks, even though the Itanium-2 has a clock rate 6X faster.

Evaluating expert-authored rules for military reasoning

Eliciting complex logical rules directly from logic-naive subject matter experts (SMEs) is a challenging knowledge capture task. We describe a large-scale experiment to evaluate tools designed to produce SME-authored rule bases. We assess the quality of the rule bases with respect to the: 1) performance on the addressed functional task (military course of action (COA) critiquing); and 2) intrinsic knowledge representation quality. In the course of this assessment, we note both strengths and weaknesses in the state of the art, and accordingly suggest some foci for future development in this important technology area.

Using hypothetical reasoning as a method for belief ascription

Abstract A key cognitive faculty that enables humans to communicate with each other is their ability to incrementally construct and use models describing the mental states of others. Every such model describing some other cognitive agent will realistically contain only a finite number of sentences in some language of thought, hence, assuming sufficiently powerful inference rules, some of its consequences will remain implicit. To make them explicit, the person holding the model could employ a kind of reasoning that can be paraphrased as ‘what would I believe if I were the other person believing everything I believe that person believes’, a strategy that can be viewed as a simulation of the other persons reasoning using the model of that person in conjunction with the reasoning abilities of the simulator. If we want to equip an artificial cognitive agent with such a simulative reasoning ability we have to cope with problems such as simulation at various levels of nesting, meta-reasoning to make implicit ag...

Implementing logic spreadsheets in less

Spreadsheets are a widespread tool for a variety of tasks, particularly in business settings. Spreadsheet users employ a form of programming that, although popular, is highly error-prone and has limited expressiveness. A promising approach to overcome these shortcomings is to augment spreadsheets with logic-based knowledge representation and reasoning (KR&R) functionality. In this paper, we present Logic Embedded in SpreadSheets (LESS), a system which integrates PowerLoom, a highly expressive logic-based KR&R system, with Microsoft (MS) Excel. The design of LESS provides different tiers of functionality that explore trade-offs between direct access to the underlying logic engine and user-friendly support for spreadsheets users. A prototype of LESS was implemented as an MS Excel add-in.

KDD-2004 workshop report link analysis and group detection (LinkKDD-2004)

In this paper we provide a summary of the workshop on Link Analysis and Group Detection (LinkKDD-2004) held in conjunction with ACM SIGKDD 2004, on August 22, Seattle, Washington, USA. We report in details about the research issues addressed in the talks and the workshop.

Caching and Consistency, a Solution in RLL-1

Caching is a term adopted from hardware technology. In the virtual memory hierarchy of a conventional computer the cache memory is a small, very fast memory located near the CPU, which is almost as fast as the CPU’s registers. This architecture assumes that a high percentage of memory accesses can be directly satisfied by the cache, and that this in total saves more time than is lost due to paging between cache and main memory.

RIPTIDE: Learning violation prediction models from boarding activity data

Part of the U.S. Coast Guards mission is to monitor vessels and their operators for compliance with a large body of safety and fisheries regulations. Recently the Coast Guard has devised a system called OPTIDE, which aims at improving operations efficiency by ranking vessels via a risk score computed from current information and aggregated past boarding observations. Ships with higher risk should be preferentially boarded, since they have higher probability of being in violation of some regulation. To improve upon OPTIDE, we developed RIPTIDE which uses machine learning to automatically learn a more fine-grained and data-driven violation prediction and ranking model from past boarding activity data. The learning problem is challenging, since the data is very unbalanced (only about 20% of all boardings actually find some violation), it has significant sampling bias, and in general the signal for predicting violations is weak. Nevertheless, our best RIPTIDE model outperforms OPTIDE by up to 86% on a ranking experiment. The main reason for this improvement comes from being able to distinguish vessels in a more fine-grained manner, which allows RIPTIDE to make winning decisions more often, even if the underlying signal is very weak. A software package implementing RIPTIDE has been developed to allow the Coast Guard to experiment with the learned models and apply them to operational data.

Estimating Violation Risk for Fisheries Regulations

The United States sets fishing regulations to sustain healthy fish populations. The overall goal of the research reported on here is to increase the efficiency of the United States Coast Guard USCG when boarding commercial fishing vessels to ensure compliance with those regulations. We discuss scoring rules that indicate whether a given vessel might be in violation of the regulations, depend on knowledge learned from historical data, and support the decision to board and inspect. We present a case study from work done in collaboration with USCG District 1 HQ in Boston.