William Rand | University of Maryland (original) (raw)

Papers by William Rand

Abstract Statistical models are widely used to explore relationships between aggregate variables ... more Abstract Statistical models are widely used to explore relationships between aggregate variables in a land-use system by inductively fitting empirical data, for example with regression models. Seldom do researchers know whether, and if so how much, the relationships thus obtained reflect the underlying mechanisms, because of uncertainties existing in the system of interests, especially when the system is a complex adaptive system. This research addresses this issue by integrating a spatial agent-based model (ABM) and ...

Participatory simulation, as described by Wilensky & Stroup (1999c), is a form of agent-based sim... more Participatory simulation, as described by Wilensky & Stroup (1999c), is a form of agent-based simulation in which multiple humans control or design individual agents in the simulation. For instance, in a participatory simulation of an ecosystem, fifty participants might each control the intake and output of one agent, such that the food web emerges from the interactions of the human-controlled agents. We argue that participatory simulation has been under-utilized outside of strictly educational contexts, and that it provides myriad benefits to designers of traditional agent-based simulations. These benefits include increased robustness of the model, increased comprehensibility of the findings, and simpler design of individual agent behaviors. To make this argument, we look to recent research such as that from crowdsourcing (von Ahn, 2005) and the reinforcement learning of autonomous agent behavior (Abbeel, 2008).

SSRN Electronic Journal, 2000

2013 Winter Simulations Conference (WSC), 2013

There has been substantial work exploring strategies, both theoretical and empirical, for selling... more There has been substantial work exploring strategies, both theoretical and empirical, for selling and buying in online auctions. However, much of this work has considered single auctions in isolation, partially because it is hard to examine multiple simultaneous auctions using traditional math modeling approaches. In reality, many auctions occur simultaneously, so there is competition not just among bidders, but also among auctions. In this paper, we use simulation to explore bidders' switching behavior between auctions for similar products. Using an empirical dataset, we first examine the distribution of switching and associated bidding behavior in real auctions. We use this data to create an agent-based model that reproduces the price process observed in the empirical data. Using this model we then explore the effects of: (1) different switching distributions, (2) the switching rule, i.e., which auction to switch to, and (3) different auction start rates. In the end, we show that in order to maximize the final price and to minimize the price disparity, auction platforms should encourage users to switch to a low-price auction that is ending soon.

Applications of evolutionary computing: EvoWorkshops 2005, EvoBIO, EvoCOMNET, EvoHOT, EvoIASP, EvoMUSART, and EvoSTOC, Lausanne, Switzerland, March 30-April 1, 2005: proceedings, 2005

Page 1. Reverse Engineering and UML: A Case Study of AuctionBot Roshan Bangera Bill Rand EECS 581... more Page 1. Reverse Engineering and UML: A Case Study of AuctionBot Roshan Bangera Bill Rand EECS 581 Prof. Sayler Page 2. Reverse Engineering and UML: A Case Study of AuctionBot Roshan Bangera and Bill Rand 2 Introduction Modeling is an important technique in science and in general since it should allow us to simplify the complexities of the world to an easily understandable form. ���A model is a simplification of reality��� (Booch, Rumbaugh, and Jacobson, 6).

ABSTRACT Agent-based modeling (ABM) has proved useful in a number of fields. Many of the early su... more ABSTRACT Agent-based modeling (ABM) has proved useful in a number of fields. Many of the early successes of ABM were due to its ability to represent the processes of a phenomenon. However, less emphasis has been placed in ABM on developing its ability to replicate spatial patterns of phenomena. In order to do that, more powerful spatial modeling techniques, like those within geographical information systems (GIS), are necessary.

Introduction lil-gp is a C language system for developing genetic programming applications based ... more Introduction lil-gp is a C language system for developing genetic programming applications based on the LISP work of John Koza at Stanford University. lil-gp evolves trees whose nodes are C function pointers, so tree evaluation is done entirely with complied code. This gives us a manyfold speed increase, allows us to handle much large problems (bigger populations, more generations), and is portable to a wide variety of platforms.

Evolution is an inherently dynamic process. The environment that a population is interacting with... more Evolution is an inherently dynamic process. The environment that a population is interacting with never remains constant. Moreover, evolution not only operates in a simple dynamic environment where aspects are fluctuating randomly, but a coevolving environment, where the the environment that a population is trying to adapt is itself evolving. Thus it is surprising that the Genetic Algorithm (GA), a computer tool which originally drew its inspiration from Darwinian evolution [16], has mainly been utilized in static environments [13].

Abstract One argument as to why the hyperplane-defined functions (hdf's) are a good testbed for t... more Abstract One argument as to why the hyperplane-defined functions (hdf's) are a good testbed for the genetic algorithm (GA) is that the hdf's are built in the same way that the GA works. In this paper we test that hypothesis in a new setting by exploring the GA on a subset of the hdf's which are dynamic---the shaky ladder hyperplane-defined functions (sl-hdf's). In doing so we gain insight into how the GA makes use of crossover during its traversal of the sl-hdf search space. We begin this paper by explaining the sl-hdf's.

ABSTRACT Several studies have discussed the importance of social interactions for product trial a... more ABSTRACT Several studies have discussed the importance of social interactions for product trial and adoption and have argued that the network structure is one of the key factors in product diffusion. However, while aggregate product adoption data is commonly observable by firms (eg, sales or subscriptions), social interactions and network data rarely is, excepting some special cases such as online communities and telecommunication networks.

1 Abstract Theoretical urban policy literature predicts the inevitability of free riding in the m... more 1 Abstract Theoretical urban policy literature predicts the inevitability of free riding in the management of common goods such as forested open space. Numerous empirical cases exist in which neighboring jurisdictions cooperate to maintain these public goods, thus challenging the expected results, yet theoretical explanations of these cases have not been fully developed.

Though recently there has been interest in examining genetic algorithms (GAs) in dynamic environm... more Though recently there has been interest in examining genetic algorithms (GAs) in dynamic environments, work still needs to be done in investigating the fundamental behavior of these algorithms in changing environments. When researching the GA in static environments, it has been useful to use test suites of functions that are designed for the GA so that the performance can be observed under systematic controlled conditions. One example of these suites is the hyperplane-defined functions (hdfs) designed by Holland [1].

Abstract Statistical models are widely used to explore relationships between aggregate variables ... more Abstract Statistical models are widely used to explore relationships between aggregate variables in a land-use system by inductively fitting empirical data, for example with regression models. Seldom do researchers know whether, and if so how much, the relationships thus obtained reflect the underlying mechanisms, because of uncertainties existing in the system of interests, especially when the system is a complex adaptive system. This research addresses this issue by integrating a spatial agent-based model (ABM) and ...

Participatory simulation, as described by Wilensky & Stroup (1999c), is a form of agent-based sim... more Participatory simulation, as described by Wilensky & Stroup (1999c), is a form of agent-based simulation in which multiple humans control or design individual agents in the simulation. For instance, in a participatory simulation of an ecosystem, fifty participants might each control the intake and output of one agent, such that the food web emerges from the interactions of the human-controlled agents. We argue that participatory simulation has been under-utilized outside of strictly educational contexts, and that it provides myriad benefits to designers of traditional agent-based simulations. These benefits include increased robustness of the model, increased comprehensibility of the findings, and simpler design of individual agent behaviors. To make this argument, we look to recent research such as that from crowdsourcing (von Ahn, 2005) and the reinforcement learning of autonomous agent behavior (Abbeel, 2008).

SSRN Electronic Journal, 2000

2013 Winter Simulations Conference (WSC), 2013

There has been substantial work exploring strategies, both theoretical and empirical, for selling... more There has been substantial work exploring strategies, both theoretical and empirical, for selling and buying in online auctions. However, much of this work has considered single auctions in isolation, partially because it is hard to examine multiple simultaneous auctions using traditional math modeling approaches. In reality, many auctions occur simultaneously, so there is competition not just among bidders, but also among auctions. In this paper, we use simulation to explore bidders' switching behavior between auctions for similar products. Using an empirical dataset, we first examine the distribution of switching and associated bidding behavior in real auctions. We use this data to create an agent-based model that reproduces the price process observed in the empirical data. Using this model we then explore the effects of: (1) different switching distributions, (2) the switching rule, i.e., which auction to switch to, and (3) different auction start rates. In the end, we show that in order to maximize the final price and to minimize the price disparity, auction platforms should encourage users to switch to a low-price auction that is ending soon.

Applications of evolutionary computing: EvoWorkshops 2005, EvoBIO, EvoCOMNET, EvoHOT, EvoIASP, EvoMUSART, and EvoSTOC, Lausanne, Switzerland, March 30-April 1, 2005: proceedings, 2005

Page 1. Reverse Engineering and UML: A Case Study of AuctionBot Roshan Bangera Bill Rand EECS 581... more Page 1. Reverse Engineering and UML: A Case Study of AuctionBot Roshan Bangera Bill Rand EECS 581 Prof. Sayler Page 2. Reverse Engineering and UML: A Case Study of AuctionBot Roshan Bangera and Bill Rand 2 Introduction Modeling is an important technique in science and in general since it should allow us to simplify the complexities of the world to an easily understandable form. ���A model is a simplification of reality��� (Booch, Rumbaugh, and Jacobson, 6).

ABSTRACT Agent-based modeling (ABM) has proved useful in a number of fields. Many of the early su... more ABSTRACT Agent-based modeling (ABM) has proved useful in a number of fields. Many of the early successes of ABM were due to its ability to represent the processes of a phenomenon. However, less emphasis has been placed in ABM on developing its ability to replicate spatial patterns of phenomena. In order to do that, more powerful spatial modeling techniques, like those within geographical information systems (GIS), are necessary.

Introduction lil-gp is a C language system for developing genetic programming applications based ... more Introduction lil-gp is a C language system for developing genetic programming applications based on the LISP work of John Koza at Stanford University. lil-gp evolves trees whose nodes are C function pointers, so tree evaluation is done entirely with complied code. This gives us a manyfold speed increase, allows us to handle much large problems (bigger populations, more generations), and is portable to a wide variety of platforms.

Evolution is an inherently dynamic process. The environment that a population is interacting with... more Evolution is an inherently dynamic process. The environment that a population is interacting with never remains constant. Moreover, evolution not only operates in a simple dynamic environment where aspects are fluctuating randomly, but a coevolving environment, where the the environment that a population is trying to adapt is itself evolving. Thus it is surprising that the Genetic Algorithm (GA), a computer tool which originally drew its inspiration from Darwinian evolution [16], has mainly been utilized in static environments [13].

Abstract One argument as to why the hyperplane-defined functions (hdf's) are a good testbed for t... more Abstract One argument as to why the hyperplane-defined functions (hdf's) are a good testbed for the genetic algorithm (GA) is that the hdf's are built in the same way that the GA works. In this paper we test that hypothesis in a new setting by exploring the GA on a subset of the hdf's which are dynamic---the shaky ladder hyperplane-defined functions (sl-hdf's). In doing so we gain insight into how the GA makes use of crossover during its traversal of the sl-hdf search space. We begin this paper by explaining the sl-hdf's.

ABSTRACT Several studies have discussed the importance of social interactions for product trial a... more ABSTRACT Several studies have discussed the importance of social interactions for product trial and adoption and have argued that the network structure is one of the key factors in product diffusion. However, while aggregate product adoption data is commonly observable by firms (eg, sales or subscriptions), social interactions and network data rarely is, excepting some special cases such as online communities and telecommunication networks.

1 Abstract Theoretical urban policy literature predicts the inevitability of free riding in the m... more 1 Abstract Theoretical urban policy literature predicts the inevitability of free riding in the management of common goods such as forested open space. Numerous empirical cases exist in which neighboring jurisdictions cooperate to maintain these public goods, thus challenging the expected results, yet theoretical explanations of these cases have not been fully developed.

Though recently there has been interest in examining genetic algorithms (GAs) in dynamic environm... more Though recently there has been interest in examining genetic algorithms (GAs) in dynamic environments, work still needs to be done in investigating the fundamental behavior of these algorithms in changing environments. When researching the GA in static environments, it has been useful to use test suites of functions that are designed for the GA so that the performance can be observed under systematic controlled conditions. One example of these suites is the hyperplane-defined functions (hdfs) designed by Holland [1].