Papers by Eswaran Subrahmanian
Springer eBooks, 2018
adaptation, distribution and reproduction in any medium or format, as long as you give appropriat... more adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
With the advent of the COVID-19 pandemic, there was the hope that data science approaches could h... more With the advent of the COVID-19 pandemic, there was the hope that data science approaches could help discover means for understanding, mitigating, and treating the disease. This manifested itself in the creation of the COVID-19 Open Research Dataset (CORD-19) which aggregated COVID-19related scientifc literature for use by the data mining community. As a group of interdisciplinary informatics researchers at NIST, we embarked on an effort to use our experience and previously developed systems to explore whether we could enhance the CORD-19 data set and facilitate its use. This effort produced a prototype scientifc informatics system that extended data curation, data repository, resource registry, term extraction and indexing systems and resulted in a repackaging of CORD-19 as a Python data package. This paper documents our efforts, provides lessons learned, and proposes a general architecture for these types of systems.
Proceedings of the ... International Conference on Engineering Design, Jul 1, 2019
In this paper we use formal tools from category theory to develop a foundation for creating and m... more In this paper we use formal tools from category theory to develop a foundation for creating and managing models in systems where knowledge is distributed across multiple representations and formats. We define a class of models which incorporate three different representations---computations, logical semantics, and data--as well as model mappings (functors) to establish relationships between them. We prove that our models support model merge operations called colimits and use these to define a methodology for model integration.
arXiv (Cornell University), Jul 13, 2023
Mathematics is a highly specialized domain with its own unique set of challenges that has seen li... more Mathematics is a highly specialized domain with its own unique set of challenges that has seen limited study in natural language processing. However, mathematics is used in a wide variety of fields and multidisciplinary research in many different domains often relies on an understanding of mathematical concepts. To aid researchers coming from other fields, we develop a prototype system for searching for and defining mathematical concepts in context, focusing on the field of category theory. This system Parmesan depends on natural language processing components including concept extraction, relation extraction, definition extraction, and entity linking. In developing this system, we show that existing techniques cannot be applied directly to the category theory domain, and suggest hybrid techniques that do perform well, though we expect the system to evolve over time. We also provide two cleaned mathematical corpora that power the prototype system, which are based on journal articles and wiki pages, respectively. The corpora have been annotated with dependency trees, lemmas, and part-of-speech tags.
Electronic proceedings in theoretical computer science, Sep 15, 2020
This paper applies operads and functorial semantics to address the problem of failure diagnosis i... more This paper applies operads and functorial semantics to address the problem of failure diagnosis in complex systems. We start with a concrete example, developing a hierarchical interaction model for the Length Scale Interferometer, a high-precision measurement system operated by the US National Institute of Standards and Technology. The model is expressed in terms of combinatorial/diagrammatic structures called port-graphs, and we explain how to extract an operad LSI from a collection of these diagrams. Next we show how functors to the operad of probabilities organize and constrain the relative probabilities of component failure in the system. Finally, we show how to extend the analysis from general component failure to specific failure modes. * * * j * j * j * j * j * j * j * j * j * j * j
Proceedings of The Royal Society A: Mathematical, Physical and Engineering Sciences, Jun 1, 2021
As the complexity and heterogeneity of a system grows, the challenge of specifying, documenting a... more As the complexity and heterogeneity of a system grows, the challenge of specifying, documenting and synthesizing correct, machine-readable designs increases dramatically. Separation of the system into manageable parts is needed to support analysis at various levels of granularity so that the system is maintainable and adaptable over its life cycle. In this paper, we argue that operads provide an effective knowledge representation to address these challenges. Formal documentation of a syntactically correct design is built up during design synthesis, guided by semantic reasoning about design effectiveness. Throughout, the ability to decompose the system into parts and reconstitute the whole is maintained. We describe recent progress in effective modelling under this paradigm and directions for future work to systematically address scalability challenges for complex system design.
Categorical Foundations for System Engineering
Springer eBooks, Nov 26, 2017
In this paper, we argue that category theory (CT), the mathematical theory of abstract processes,... more In this paper, we argue that category theory (CT), the mathematical theory of abstract processes, could provide a concrete formal foundation for the study and practice of systems engineering. To provide some evidence for this claim, we trace the classic V-model of systems engineering, stopping along the way to (a) introduce elements of CT and (b) show how these might apply in a variety of systems engineering contexts.
In this paper, we make the case that ICT projects in the developed and developing world often lea... more In this paper, we make the case that ICT projects in the developed and developing world often lead to partial or total failures due to the incomplete assessment of the problem being solved and the metrics used to evaluate solutions. While in the developed world the success of ICT solutions are often determined by the market, with available infrastructure and market mechanisms, in the developing world this ecosystem does not exist thus requiring an understanding of the ecosystem in which ICT solutions are to be applied. Using literature from the design space, and experiences in ICT for development, we elaborate the dimensions of design such as incorporation of stakeholders, incentive structures, and design participation that are critical to successful deployment. We examine some successes and failures in product/solution development in the ICT area to identify the dimensions of good design incorporated by these products and services. With the perspective that ICT for sustainable development issues are ill-structured and "wicked problems" that have to incorporate all the defined dimensions of design, we propose a model of product and service identification and development that is based on insights from asynchronous computational agent problem solving. We claim that new methods such as the one proposed need to be identified, developed and tested for their effectiveness in the development of products and services that satisfy the needs of human development.
Flexible extraction of practical knowledge from bridge databases
ABSTRACT
Computers in Industry, Nov 1, 2019
Process plans provide a structure for 1) identifying the tasks involved in a given process, 2) th... more Process plans provide a structure for 1) identifying the tasks involved in a given process, 2) the resources needed to accomplish them, and 3) a variety of relationships and constraints between these. This information guides important operational decisions across various organizational levels, from the factory floor to the global supply chain. Efficient use of this information requires a concrete analytical model that can be easily represented in digital form. In this paper we present a modeling framework for process plans based on a branch of mathematics called category theory (CT). Specifically, string diagrams provide an intuitive yet precise graphical syntax for describing symmetric monoidal categories (SMCs), mathematical structures which support serial and parallel composition. Ideal for process representation, these structures also support a powerful mathematical toolkit. Here we use these tools to analyze the relationship between different levels of abstraction in process planning hierarchy. We also model some spatiotemporal aspects of planning, dynamic decision-making inside the process hierarchy and the updating procedure in the event of errors or extensions in the production line.
Flexible Extracation of Practical Knowledge from Bridge Databases
Computing in Civil Engineering, 1994
Modeling the Internet of Things and People: A Foundational Approach
Proceedings of the Design Society, Jun 19, 2023
The modularity of components has enhanced the ability to create IoT systems by composing them fro... more The modularity of components has enhanced the ability to create IoT systems by composing them from off the shelf. However, the breadth of technological choices and capabilities of component devices has made designing these systems harder to select, compose, implement and test, especially for dynamic systems. In this paper, we adopt formal tools from category theory (CT), a branch of mathematics whose central tenet is compositionality, to generate models for IoT systems. More specifically, we introduce a port-graph operad to represent the architectural designs of IoT systems. We use presheaf categories to construct generic IoT schemas to support modularity. Given this information, we briefly describe its relationship to control strategies of dynamical systems that model the interaction of components. Our approach balances genericity and specificity, providing interlinked schematic representations of system architecture and component representation.
Category Theory
Springer eBooks, 2023
This study analyzes the workflow and implementation of Electronic Health Record (EHR) systems acr... more This study analyzes the workflow and implementation of Electronic Health Record (EHR) systems across different functions in small physician offices. We characterize the differences in the offices with different levels of computerization in terms of workflow, sources of time delay, and barriers to the support of the entire workflow through the use of EHR systems. Design: The study was based on a combination of questionnaires, interviews, in situ observations and data collection efforts. The study was performed at 13 small practice physician offices in the Baltimore and New York areas. The offices included both primary care providers (4) and specialty providers (9). A total of 50 person-hours were spent with the staff to observe and capture the present work practice. This study was not intended to be a full-scale time-and-motion study with precise measurements, but to provide an overview of the potential sources of delays while performing office tasks. Method: This study follows an interpretive model of case studies rather than a large sample statistical survey of practices. Workflow maps were created based on the aggregated data from the offices to identify time-consuming tasks. This information was used to compare and contrast the workflow maps for offices with and without EHR implementation. The workflow maps for specialist and primary care physicians were studied to understand the differences in workflow complexity that may be a factor in the differential adoption of EHR systems by the two types of offices. Results: The results from the study show that specialty physicians are more favorable towards adopting EHR systems than primary care physicians. The results further indicate that while primary care physicians favored introducing EHR systems in their offices, the barriers to do so were greater than those encountered by specialists. This can be attributed to the complex workflows that exist in the primary care physician offices, leading to non-standardized workflow structures and practices. The results suggest that primary care physicians would benefit more from EHR systems if they could interact with external entities through computer-based transactions. Presently, non-standard ways of interaction with external entities and inadequate computerization are the primary barriers to moving away from paper-based systems. Without being able to move away from a paper-based system, an EHR system becomes an extra burden and results in low rates of adoption. Conclusions: Significant variations were observed in workflow characteristics between primary care and specialty physicians that could be attributed to the variety and multiplicity of external interactions. It is easier to adopt an off-the shelf system if the variety and number of interactions are constrained. The manual nature of the tasks in non-EHR-based offices results in considerable waste of time and increase in cost. Any move to an EHR system should hence include a strategy for communicating with the outside world.
This is the first series of papers critiquing the Unified Modeling Language (UML) and proposing a... more This is the first series of papers critiquing the Unified Modeling Language (UML) and proposing an alternative modeling language based on a branch of mathematics called category theory (CT). We argue that modeling in CT provides a richer and more precise formal structure than UML, while still supporting many of the intuitive and diagrammatic features which engineers appreciate. Deep ties connect CT to formal logic, physics, and computer science, helping to connect our high-level models to simulation and implementation. In this paper, we focus on modeling UML's structural component, as exemplified by the class diagram using CT models. To do this, first, we introduce CT and walk through the constructs of the class diagram one by one. Later, we show how UML class models are implemented through standard CT constructions. As we do, we show that some UML constructs are redundant while others smuggle in assumptions that should be made explicit. Thus, we argue that the use of CT in information modeling could make our models less ambiguous, more precise, and more formal.
Compositional Models for Complex Systems
Elsevier eBooks, 2019
Abstract In this chapter, we argue for the use of representations from category theory to support... more Abstract In this chapter, we argue for the use of representations from category theory to support better models for complex systems, and provide an example of what such an application might look like. Our approach rests on the well-known observation that complex system design is a fundamentally recursive discipline, which is formalized in computer science using structures called algebras, coalgebras, and operads, mathematical structures that are intimately linked to labeled tree representations. We then develop two small examples to demonstrate the utility of this approach. The first defines a logical semantics of contracts to organize requirements at different scales in hierarchical systems. The second concerns the integration of artificial intelligence models into a preexisting human-driven process.
IEEE Computer, Dec 1, 2021
Engineering safe and secure cyber-physical systems requires system engineers to develop and maint... more Engineering safe and secure cyber-physical systems requires system engineers to develop and maintain a number of model views, both dynamic and static, which can be seen as algebras. We posit that verifying the composition of requirement, behavioral, and architectural models using category theory gives rise to a strictly compositional interpretation of cyber-physical systems theory, which can assist in the modeling and analysis of safety-critical cyber-physical systems. Recently all these three areas of control [10], contracts [4], and co-design [11, chapter 4] have been described, generalized, and unified with fun
Journal of the Indian Institute of Science, Oct 1, 2020
Introduction According to a report by the Centers for Medicare and Medicaid Services [CMS] (https... more Introduction According to a report by the Centers for Medicare and Medicaid Services [CMS] (https ://www. cms.gov), the United States spent nearly $3.6 trillion dollars on health care in 2018. This amount is about 17.7% of the nation's GDP. Assuming a steady increase, this expenditure would be nearly $4 trillion dollars in 2020. Hence, it is important to understand the various trends that will affect the future of health care. The journey to future health care will involve advances mainly in three dimensions (see Fig. 1): (1) advances in health care technology; (2) advances in health care delivery; and (3) advances in computer science and information technology. Advances in health care technology include
As we extend the reach of the Internet through sensing and automation, networked interactions bec... more As we extend the reach of the Internet through sensing and automation, networked interactions become more critical and safety and security require improved means for designing and testing system components, platforms and services. A tremendous amount of messy, heterogeneous data passes through such networks; integrating and applying that data so that it can be used for seamless interoperability and higher level reasoning requires conversion into some higher abstraction. This necessitates a new foundation for information modeling and model management, broadly construed. Such a foundation must support model construction in a wide range of formalism as well as capabilities for integrating models across different formalisms. We must also allow for easy model evolution, simplifying both iterative design and long-term maintenance. In this paper, we argue that category theory, a branch of abstract mathematics, provides a firm conceptual foundation for information modeling which already meets most of these criteria. We close with a discussion of some challenges for exploiting category theory in applied contexts. CCS Concepts • Computing methodologies Modeling methodologies. • Applied computing Mathematics and statistics. • Applied computing Information integration and interoperability.
Uploads
Papers by Eswaran Subrahmanian