Archive for January, 2012

ABMs: representation, coherence, balance

Monday, January 30th, 2012

O’Sullivan’s article is a rather critical account of ABMs. The article states the issue of highly funded models, which are too sensitive to reveal the outcomes of or are too complex to be explained in journal articles (544). State-of-the-art findings will reach a specific audience, not the audience that they were intended for. Thus, if ABMs are social agents, representing social issues, we have one serious limitation. How will transparency, availability, and clarity be attained? Perhaps we should strive for balance in models between the relationship of agents with space, and how and where those agents are represented (545).

I found it difficult not to get lost in the definitions of ABMs. In particular, their accuracy, validity, and lucidity. Bonabeau differentiates ABMs from market models with advancing game theory, by taking the focus off of the ‘theory’ part. On the other hand, O’Sullivan finds ABMs to be simple and abstract, effective for researching theory implications. He goes on further to state that ABMs, as they stand, “cannot establish the truth of those theories” (546). How then, can the truth of those theories be established? Should we be concerned with the idea of theory? Or restructuring what we deem that a model is all together? The issues with regards to the way space and time are represented are being explored and that’s definitely a good thing. Despite all the setbacks, there is much potential in ABMs as they are exploring numerous fields. As long as limitations diminish, there is hope.

-henry miller

ABMs and MBMs

Monday, January 30th, 2012

The focus on Bonabeau’s article here is of agent-based models (ABMs) and market-based models (MBMs). A main difference between an ABM and an MBM is that the former focuses on individual behaviour while the latter deals with collective behaviour. Furthermore, ABMs are driven by the bottom-up approach (focus is placed on the individuals), in contrast to the top-down approach (focus is on the collective) of the classic MBMs.

“ABM captures that emergent phenomenon in a natural way” (7282). Is this much different than a general equilibrium MBM? The MBMs that are largely based on the neo-classical economics are founded on similar ideals of natural systems, such as laissez-faire economics and Adam Smith’s infamous ‘invisible hand’. Referring to another model, Bonabeau states that “each agent acts individually but has perfect knowledge of how many users there are in the population” (7287). Whenever ‘perfect’ is utilized, it reminds me of the classical economic market systems where perfect competition, perfect information, and full employment are all assumptions made in assessing a market scenario with a standard economic model.

Bonabeau is at times too eager and possibly even blinded by his excitement of his tool: “ABM is perfect not just for operational risk in financial institution but for modeling risk in general” (7285). It is problematic to be this certain about a tool that is not fully understood, which he actually takes note of: if ABM is introduced in the market and is unsuccessful, potentially harming individuals with the predictions, or rather understanding (as the article emphasizes) of a situation, then the overestimated tool may do more damage than good. However, the article does mention this, but does not go into detail explaining why “agents behave in a way that is still poorly understood” (7284). Perhaps the statement “AMBs are more of a mindset than a technology” (7280)alludes to this problem. Can we change the modern market system with ABMs if we are aware of their positive and negative implications? Is it possible to create economic software agents that do not simply explain human economic behaviour?

-henry miller

ABMs are hard!

Monday, January 30th, 2012

Bonabeau (2002) articulates the common danger of “improper use of ABM” (7280), regarding the simple technique in ABM creation coupled with the need for conceptual rigour. I don’t think he really explains himself on what he means by “conceptually deep” (7280) throughout the article, and I can see two ways for it to be taken.

With the attempted replication of agent interactions it can be assumed that lots of data is required for the model to be held as valid. Putting value on heterogeneity, individual data can be associated to agents, and there can be multiple different agents in the model doing different things. As agents can exhibit “learning and adaptation” (7281), this needs to be incorporated into the model along with agent rationality and some knowledge of the environment, or adherences to spatial parameters. Modellers are attempting to simulate real occurrences, and we know that human behaviour is incredibly difficult to predict and account for.

Another way I see models as being conceptually deep is in the analysis stage, post-programming. Bonabeau queries “what constitutes an explanation of an observed social phenomenon?” (7281). ABMs capture emergent phenomena, but taking the step to explain this emergent phenomena may prove more challenging for social scientists. With ABM we can make things happen from the bottom-up, and then need to seek reasoning for the phenomena we create, which may not always be evident.

Bonabeau, Eric. “Agent-based Modeling: Methods and Techniques for Simulating Human Systems.” Proceedings of the National Academy of Sciences of the United States of America. 99.10 (2002): 7280-7287. Print.

-sidewalk ballet

How to Define Agent-Based Modeling in Human System Study?

Monday, January 30th, 2012

Agent-based modeling (ABM) has been widely utilized in social science as a powerful modeling tool in the past few years. However, it still remains difficult to conclude an accurate definition of ABM. When it comes to the study of human systems, the situation becomes more complicated due to the uncertainties in human society.

In this paper, Bonabeau conclude that ABM provides three important benefits for simulating human systems. First, it presents a natural description of the study target. Second, it is quite flexible and can be easily adapted for different dimension with the ability to learn and evolve by itself. Finally, ABM captures emergent, which is quite challenging and important in study of human systems. All these features have conceptualized ABM as a necessary tool to handle the complexity in human system studies.

I tend to define ABM as a collection of modeling methodologies, rather than a group of theories. ABM contains a wide body of disciplines, and it seems the number of disciplines will continue increasing. It relies on the methodologies and techniques of statistics, pattern recognition, reinforcement learning, control theories, system identification, game theories, and so on. However, ABM distinguishes itself from the mentioned disciplines by more emphasis on practical usage, not on the solid theory background. For example, ABM can model auction activities without the proof of Nash Equation’s existence, which is not acceptable in game theory studies. Moreover, ABM may utilize reinforcement learning as a tool for self-learning, without any theoretical proof of whether it can converge to the satisfying result or not. Therefore, I think ABM as a collection of practical modeling methodologies, and it will change its definition with different applications.

–cyberinfrastructure

Modeling the Individual for the Whole: Apocalypse Group Dynamics in ABM

Monday, January 30th, 2012

One of the major issues that Bonabeau brings up in his conclusion is the notion that ABMs model a system by simulating the actions of the individual units of analysis and not the group as a whole on the aggregate level. Because of the individual simulation of agents, this process is very computation-intensive and leads to high hardware costs or investment in cloud-computing infrastructure.  He notes that while aggregate-level analysis could be done with just a few equations, it is more complicated and time-consuming to describe individual units.

In the example of modelling a Zombie Apocalypse, I would like to see how the actions of the individual agents affect the outcome for the group as a whole.  As commonly portrayed in Zombie Sci-Fi, even a close encounter with an infected agent can have dire consequences for the entire group that the affected agent belongs to, due to the slow nature of death by the pathogen, and the way in which it manifests itself in its recently dead hosts, thus putting the entire group at risk from an un-noticed bite victim.  I would like to see how the agents would adapt as a group, if one were to add in the preference for both non-infected and infected agents to remain in close proximity to like agents.

 

-Rsmithlal

Using ABM to Model a Zombie Apocalypse

Monday, January 30th, 2012

According to O’Sullivan, the field of Pedestrian models is one of the up and coming areas of interest in what he calls “a locally specific agent-based approach” (O’Sullivan, 543).  In his section on “mid-range regionally or locally specific models”, he states that recent work has been done to simulate crowd-control of large groups of pedestrians in a panicked situation.  I would like to take this approach much further and propose a what-if scenario in which ABM was used to model what would happen in a situation where a real-life pathogenic Zombie outbreak were to occur.

Some suggested parameters to add to the model would be the underlying city infrastructure, locations of food, water and first-aid, locations of weapons and ammunition as well as areas that can be considered safe to occupy.  One of the key features that would have to be programed in the agent’s ability would be non-infected vs infected and how individuals vs groups would respond to either individuals or groups of Zombies. In O’Sullivan’s example, he notes that while the environments that the agents occupy may be complex, the agents themselves are not complex.  Ideally, it would be a simple matter of the flight or flight response, with added thresholds for when the risk of danger is outweighed by the need to venture outside to scavenge materials.

I wish that I understood more about building these sorts of models, as this is a theoretical situation that I would be thrilled to be able to simulate.

-Rsmithlal

Utilizing ABM in GIS Research

Monday, January 30th, 2012

O’Sullivan gives a very good survey about the agent-based models (ABM) in spatial science research. He begins with different definitions of ABM, analyzing their advantages and disadvantages. Then he categorizes ABM applications into three types, with respect to their degree of complexity. As mentioned by Bonabeau, ABM provides an efficient approach to describe the complex systems. O’Sullivan illustrates this point in the paper by delineating spatiotemporal and social ABM representation methodologies. In geography research, ABM provides a powerful tool for modeling geospatial information on computers. However, the challenges should not be overlooked. As ABM represents geospatial information at individual level, the complexity and model verification are becoming more difficult with the increase of agent numbers.

By modeling interesting entities as agencies, ABM reviews the relationship and interactions between these agencies. Sometimes, the study targets in GIS research are in a great number and their relationship can be very complicated for statistical modeling. But ABM can achieve that by modeling each entity or each attribute of the study target, providing detailed models about the intricate phenomenon.

I propose that ABM should be integrated with other geospatial analytical methods in GIS research. By viewing a large number of agencies at aggregate level, we can find several interesting discoveries that cannot be reflected by studying the interactions between the agencies. For example, geospatial statistics can study climate changes at global level, by utilizing geospatial information provided by a large body of agencies which contribute to the climate change. Therefore, utilizing other analyzing tools with ABM can help us in GIS research.

–cyberinfrastructure

ABMs as a Tool?

Sunday, January 29th, 2012

In O’Sullivan’s description of simple abstract models, he mentions Epstein and Axtell’s use of the term ‘generative’. It is discussed initially as a “new approach to social science” whereby if a model “replicates observed regularities in the real world”(542) the researcher can claim to have “explained the phenomenon”(542). Although this is mentioned in the context of social science, it strikes me as being completely against the use of the word ‘explanation’ when it is applied to science. To me, a scientific explanation involves being able to not only list factors that create the phenomenon and quantifying them but also understanding the interactions between them. In this case, ABMs are being used as a tool rather than a science and involve more trial and error button pushing than anything else. If one sets the parameters of the model in many different ways for each simulation and one happens upon the right combination of parameters, one has managed to explain a phenomenon. Thus, in my opinion, it is necessary to ensure one fully understands all the factors influencing a system and how these factors interact with one another before one can say they have explained the system even when an ABM produces results matching the system as seen in reality.

-Outdoor Addict

 

A Missing Perspective on ABMs: The Developing World

Sunday, January 29th, 2012

Bonabeau speaks of four main areas of application in which ABMs can be used: flows, markets, organizations and diffusion and organizes his article around these four applications with examples for each. What caught my attention was that most of the examples drawn for these applications relate to the developed world. Technological innovation has clearly been far more rapid and widespread in the developed world which could account for the current uses of ABMs being from developed countries. Granted, the applications of ABMs in developed countries could just as easily be used in developing countries as these face the same issues as developed countries with respect to situations such as traffic jams, evacuation from crowded areas, transit and stock markets among others.

What I would like to see more with ABMs is use of ABMs in developing countries to simulate the way some situations may impact these countries in different ways than developed countries. One example of this can be seen through the application on ABMs by diffusion. Bonabeau describes diffusion as an application for ABMs where “people are influenced by their social context” (7285). The diffusion of education and knowledge in space in developed versus undeveloped countries could be interesting to examine as the processes for this could be very different in these countries and could be performed by very different processes dependent on many factors such as the spread and use of technology such as computers, cell phones or social networks that may exist online or face to face. In a broader context, how might ABMs be applied in development scenarios in developing countries?

Bonabeau, Eric. “Agent-based Modeling: Methods and Techniques for Simulating Human Systems.” Proceedings of the National Academy of Sciences of the United States of America. 99.10 (2002): 7280-7287. Print.

-Outdoor Addict

Uncertainty in ABM

Sunday, January 29th, 2012

O’Sullivan posits that the complex nature of ABMs violates “one of the most common tenets of practical science, the imperative to prefer simplicity over elaboration” (p. 546). As stated in the post “ABM and Toolmaking,” many issues arise due to this complexity. It is difficult to image, however, a simplified model of human processes; there are so many important variables to take into consideration. But in developing these models—especially when they are used not just to understand a phenomena but also to predict—what happens when the predictions are incorrect? Perhaps this is an issue that Bonabeau does not delve into enough: it is possible that the over-simplification of a system or the inability to consider enough variables can lead to error and uncertainty. In class we discussed the Turcot interchange and the designing of the freeway system in general. Computer models, which may have not taken into consideration enough transportation demand and urban growth variables, may have let to inappropriate policy and planning decisions.

Again, this may be a problem that can be solved through technological advancements. Revisiting the freeway example, we can now model how expanded roads quickly reach capacity. But maybe there is another issue at play, that of scale. How can a model that represents the actions of agents be simplified and, therefore, more accurate? Bonabeau uses the very small-scale example of the fire escape simulation to assert the benefits of ABM. In this example, while there are relatively few and homogenous actors who all have a single aim, the idea to construct a column in front of the exit would likely not have been arrived at without the aid of ABM. An alternative way of problem solving and perceiving a process, in other words, was enabled. Therefore, perhaps the effectiveness of ABM lies in fully understanding its limitations.

According to O’Sullivan, “while simple, abstract models can be useful for exploring the implications of theories under particular assumptions, they cannot establish the truth of those theories” (p. 546). So, it’s great that we can now determine the optimal location of a column, but the current nature of ABM will mean that fully understanding a complex social phenomena will be riddled with uncertainty. Keeping this important aspect in mind will arguably be key to the success and future development of ABM.

– jeremy

ABM and Toolmaking

Sunday, January 29th, 2012

As illustrated by O’Sullivan, individuals in an ABM are governed by a set of rules in order to provide a natural representation of a phenomena. I am intrigued by the notion of having such a flexible system, which allows agents to exhibit behaviours and reactions that differ from their counterparts. This reflects real-life situations, as the outcome of an event, for example, can completely depend on the actions of a single individual, who is influenced by their setting and social context and vice-versa.

Our in-class exercise of modeling the users of McGill’s outdoor walkways, however, revealed the incredibly complex nature of agent-based modeling. In attempting to represent how students, tourists, cyclists and vehicles use space, we quickly discovered that there are a wide range of users who use the space very differently, and also respond to certain events in very different ways.

This complexity brings up issues surrounding data collection/storage/usage limitations, which have arguably rendered the popularity of ABMs to be low in the geographic community to date. As noted by sah, I think that this issue has not been given enough attention regarding access to modeling capabilities. More importantly, however, because AMB is still in its infancy, I think that the way in which ABMs simulate human systems reinforces the notion of GIS as tool-making, a process whereby representations are constantly being improved upon. While GIS may currently struggle to represent processes, technological advancements—as O’Sullivan briefly illustrates—will perhaps enable ABMs (and GIS) to better incorporate the human element through increased public participation, for instance.

– jeremy

Emergent Phenomenon Revisited

Sunday, January 29th, 2012

I think one of the most interesting aspects of our exploration of ABMs isn’t the models themselves but the concept of emergence which lies at the heart of this whole methodology. Thinking of systems as the whole of a good many moving parts implies a startling paradigm shift whereby many systems might simply arrive at a destination through no prior planning or intentionality. Instead, these systems create complex patterns or phenomenon simply due to a variety of independent agents going about their business.

So far, we’ve thought about emergence in terms of banks or traffic which allows for such an explanation without too much hesitation (although, as a bank manager, I would certainly like to think I have much more control over the functioning of my business). But what about when we apply the idea of emergent phenomenon to more natural science-based systems?

I realize both of our authors write about ABM – for the most part – as a new tool for social scientists. Eric Bonabeau, for example, appears more likely to discuss crowd panic (7282) or the role of ABM in social sciences (7287) than ant hill dynamics or starlings (cool video, by the way). Yet many of the examples of ABMs that we saw in class (such as Boids) could just as easily involve natural systems. One might easily consider biological systems as models for how complex behavior can stem uncreated from far simpler behaviors such as the chemistry of carbon compounds.

Our world is filled with both natural and sociological situations that display patterns of emergence, as both Bonabeau, O’Sullivan and Peter have pointed out. This idea – emergence – provides a useful paradigm for understanding and exploring this phenomenon wherever it may occur. ABMs may just be a part of this exploration.
–ClimateNYC

ABM – what else is there

Sunday, January 29th, 2012

The papers discussed bring up some very important issues about ABMs – how useful are they when models are too simple, how can we extract real causal relationships when they become more complex and start to mirror the real world. Problems of equifinality and how to evaluate ABMs are also important – how are we to verify and validate an ABM that we use for predictive purposes? Even though there are many problems concerning the use and interpretation of ABMs, I think it is still important to acknowledge that these are very cost-effective and quick forms of social experimentation that do not require large amounts of time, manpower and money to perform. There was perhaps one problem that we may see with models such as Schelling’s segregation model – the fact that his model had a final steady state. It seems that if an ABM is too simple, it is very possible to end up with a final steady state, such as with the case of segregation. This is hardly the case, depending on the temporal scale one is looking at, and the accuracy of the programming of agents. Most systems in the world tend to be in flux rather than unchanging – this is what makes the world complicated. Therefore, if conclusions of steady states arise from ABMs, it is perhaps better to use a more complicated model. There tend to always be some kind of exogenous factors that will affect the output of a real world system, and this is what makes ABMs so hard to work with. However, does that mean that Schelling’s model should take into account income levels, land values, rent values, available services etc.? I do not think so, as it would convolute the question and focus of the model away from ‘individual preferences for like individuals’. Having too many variables in a model just serves to blur away any possible causality. This is a problem with all sciences, but especially problematic for ABMs since they tend to deal in complexity rather than simplicity. Some previous comments here have noted the high demand ABMs have for computational power. This is increasingly becoming less of an issue, and becoming more of a data transfer, and data structure problem in my opinion. Actual processing power will not be the limit in the future, only our own data and programming structures with which we create ABMs.

Another very limiting factor to ABMs is calculating error and uncertainty. How should this be done, especially when used for ‘predicting’ or ‘forecasting’, and when we cannot truly model every single possible action of real-life agents? I think this is one of the problems of ABMs that holds it back from mainstream science or even GIS. Whereas in say, hyperspectral imaging, you can attribute your error to the sensor and other factors such as your calibration and correction, in ABMs it would be difficult to assign any sort of error value to conclusions, especially those that do not have a real-world comparison.

Finally, I would like to draw us to the question of: are there alternatives to ABMs? I believe the answer is no. Social experimentation involving large numbers of individuals is too difficult to control in the real world, and much more consuming in terms of time and money. A large-scale real life social experiment is just not as efficient. Additionally, ABMs have the important feature of being able to be re-run very easily – but real-life social experiments cannot just be ‘reset’, especially when the researcher doesn’t memory of the previous experiment to influence agent behaviour.

-Peck

A Bit More On The Appropriateness/Drawbacks of ABM

Sunday, January 29th, 2012

I know a couple of folks have posted on this topic, but I wanted to add my two cents given that both of our authors for last week treat the drawbacks of ABMs in some detail. In particular, I’m interested in David O’Sullivan’s idea that simple models are necessary in science in order to arrive at understandable explanations of what’s taking place within the model or with an emergent phenomenon (546). While understandable as a scientific paradigm, I think this approach may explain the sentiment captured in “Sidewalk Ballet’s” post and the subsequent debate on this blog about whether ABMs can truly capture/represent life – particularly in geographic terms.

I don’t completely agree with “GIS Funa” that ABMs should only be used as a means for “breaking down” complex phenomenon. While O’Sullivan appears to accept that ABMs often are used this way, he, himself, writes that while simple ABMs might be useful for exploring theories under particular assumptions, they could never be used to “establish the truth of those theories” (546).  He adds that their logic would never be more convincing than other “rhetorical device[s]” if this was the only manner they were used (546). He goes on to conclude that the challenge for modellers is to find more sophisticated ways in which to use ABMs (just as modelers working with other types of models have done).

While I believe ABMs might always be a bit soft in explaining potentially complex individual actions such as irrational behavior, subjective choices or other complex psychologies (as Eric Bonabeau suggests on p7287), they can approximate reality and account for geographic space. Furthermore, these approximations can be extremely useful in trying to better understand complex, emergent phenomenon, flows, or thresholds/state changes in a system. So, we must instead learn how to use ABMs for purposes where these strong suits can best be harnessed – while remaining aware of any limitations.

–ClimateNYC

ABM’s and the Ballet

Sunday, January 29th, 2012

The Sidewalk Ballet speaks very critically of ABM’s: “Abstract ABMs disregard any detail of real world situations […]Things occurring in one space cannot be blindly applied to a different one without acknowledging the different factors which comprise and inhabit the space[…] then what is their use in GIScience?”

I will argue that very few ABM’s will claim to explain the whole issue being examined. I do not believe that the people responsible for the ant and the sugar model are attempting to explain income inequality. The metaphor there or the main purpose behind ABM’s in general is to simplify a certain problem to gain a better understanding.

If I were to, for example attempt at writing a piece of music, I would not write everything down at once—or if I were to decide deconstruct and learn how to play and conduct, let’s say Tchaikovsky’s Swan Lake ballet Op.20 written in 1875, I would not tackle the whole song at once. I would break down the components. First I might figure out what key the song was written in. I would then have a better understanding of what my options may be when choosing the notes throughout the song. From there I might teach myself different chords or scales that are required to play this song. Sure, if I blindly apply scales outside of the proper key into the song, my rendition might not sound good, but slowly, by gathering more knowledge of the problem I become closer to learning the piece as a whole. By understanding the key, time signature and tempo, I could then change the way I conduct the song. By changing one of these components, the mood of the song (outcome) can drastically be altered.

I’m not totally in favour of ABM’s, and I do understand some of the drawbacks, but I do think that they offer a way for us to break down complex phenomenon. As mentioned in class, the IPCC attempt, time after time, to model the impacts of GHG’s on climate change. People have been extremely critical of these models, but I believe that the point here is that models aid in the understanding of trends and patterns. Many miss the point and are very focused on proving these predictions (I will use this word gingerly) wrong.  By being able to control variables, much like key or time signature, researchers have the opportunity to forecast and understand possible outcomes with very little risk. By being able to change desired variables, we now have the ability to see how each one impacts the system. As basic as the variable sliding bars seen in class appear, they still represent an extremely powerful tool.

http://youtu.be/_Uyg24hOiog

Great movie, great ballet.

Andrew “GIS” Funa

 

Re: GIS:ABMs (O’Sullivan, 2008)

Sunday, January 29th, 2012

I particularly liked how O’Sullivan’s introduced the various types of ABMs by separating them into three categories depending on how realistic they are because it reminds me that, although we can create incredibly complex ABMs that resemble reality very closely, the value in “simple abstract models as thoughts experiments” (542) should not be underestimated. Bearing in mind that “complicated models may remain just as baffling as the world they purport to represent” (546), perhaps for many research questions, extremely realistic stimulations are not necessary. Simple models that explore the interactions between only a few theories can no doubt shed new light on problems even if the stimulated scenarios are not observed in reality. Thus, building ABMs to stimulate thought experiments could prove to be a useful tool at the exploratory stages of research and theory building.

The concept of equifinality and model verification also got me thinking. In complex and flexible systems, isn’t it more common to be able to reach a certain outcome through different means than through only one means? I think learning to accept the fact that many models may be “valid” and evaluating model outcomes in terms of “… the trajectory by which those outcomes are reached” (546-547) must go hand-in-hand. For instance, in a game of chess, the same final outcome (e.g. checkmate) is reached through thousands of different sets of moves. Thus, when comparing two highly skilled chess players, it is much more convincing to evaluate how each player executes his moves than to see whether or not he/she can deliver a checkmate.

-Ally_Nash

Agent-based models are cultivators

Saturday, January 28th, 2012

Bonabeau’s article certainly gives us a good introduction to the potential of agent-based modeling and the wide-reaching social phenomenon that it is able to explore. The key contribution that ABM offers social science is the ability to study an issue from bottom-up by looking at interactions between agents rather than overall processes produced by agent. The concept of “growing” an explanation to social phenomenon is both catchy and intriguing.

I especially liked the last example in Bonabeau’s paper. It showed how an ABM is capable of incorporating the fact that each individual is situated in the social world and that we can most influenced by the people we know both in terms of our behaviors and our location (http://gizmodo.com/5879504/how-your-friends-locations-give-yours-away-online). In an age where social network sites allow us to communicate with more people than ever before, ABM can be powerful in studying the adoption of new ideas, dissemination of information, and power/influence.

However, there were two points which I thought deserved further explanation. First, when arguing for the flexibility exhibited in an ABM, the author points to its “… ability to change levels of description and aggregation: one can easily play with aggregate agents, sub-groups of agents, and single agents, with different levels of description coexisting in a given model” (7281). I think this is an attractive feature that warrants more description regarding how groups are created, whether a single agent keeps its description when aggregated with other agents, and can an agent move dynamically in/out of a group? Secondly, Bonabeau mentions the ability for ABM to capture “individual behaviors [that] exhibit memory… learning and adaptation” (7281), but fails to mention the type or the complexity of knowledge that can be learnt by agents (are logical inferences possible?). It would have been interesting to see an example of how this artificial intelligence plays out in an ABM and a brief discussion about the power and recent developments in the capability of this type of algorithms.

-Ally_Nash

Necessity of ABMs?

Saturday, January 28th, 2012

Agent-based models are an interesting concept, with much potential, and as O’Sullivan describes, many limitations.  As exciting as they are, they have not gained the popularity in geography one may have expected, either.  Several reasons for this stood out to me amid the papers and class discussion.

First, agent-based models of increasing complexity are expensive, but if we want them to put out detailed results, they require this commitment of time and money to collect and input extensive amounts of data.  How many companies can afford the required commitment?

Second, as O’Sullivan mentioned, ABMs “frequently violate one of the most common tenets of practical science, the imperative to prefer simplicity over elaboration”.  This, combined with the fact that complex results can sometimes be just as hard to fathom as the complex data entered into the models, can be off-putting.

Third, and I would argue, most relevant, is the current availability of real-time data.  In class, we discussed how a lot of the results ABMs are attempting to discover can currently be monitored in real-time.  If we can have access to real people outputting results, as opposed to agents endowed with real qualities, why would we not choose the former?  Surely they can best model the complexities and counter-intuitive nature of real life?

All that being said, ABMs seem incredibly exciting, and an interesting way to model problems that we don’t have answers and data for already.  With continued improvements into the future, they seem like a technology to watch, maybe for emergence in unexpected ways.

-sah

O’Sullivan, David. “Geographical Information Science: Agent-Based Models.” Progress in Human Geography. 32.4 (2008): 541-550. Print.

O’Sullivan reading and Conceptualizing Agents and Environments

Saturday, January 28th, 2012

I would like to respond and build on to sidewalk ballet’s post about the contextual limitations of abstract Agent Based Models (ABMs). While your critique may be levelled at abstract ABMs specifically, I believe that ABMs are capable of capturing the complex specificities of any local area. Coming at a computational cost, I would imagine users of ABM would typically tailor their models to accurately reflect their situation. Another thing to consider is that an ABM’s environment does not need to represent a physical location. ABM can be used with environments that are very common to many agents, such as time or virtual environments like the stock market. I think of it as a way to model simultaneous decision-making, game theorying. In these types of non-place specific environments, more generalized ABMs may still be appropriate.

An important reminder is that agents need not be individuals as well. Frankly, I find this a more challenging concept for ABM. O’Sullivan touches on this point as well; how do we represent agents that are not individuals and who are also mobile? Is it ever appropriate to assume an entire family is one agent when a census tracks movement of the population? Can we assume a pride of lions in the savannah is one agent since it is clearly led by the dominant male? Tackling these issues of generalization and how to represent their movements allows for reduced computation cost and a scaling up of these models.

O’Sullivan, David. (2008). “Geographical information science: agent-based models.” Progress in Human Geography, 32(4) 541-550.

– Madskiier_JWong

Bonabeau reading and acknowledging limitations of Agent Based Modeling

Saturday, January 28th, 2012

Bonabeau’s article goes to great lengths to illustrate the advantages of Agent Based Modeling (ABM). He provides a quick overview of the approach which consists of agents that independently make decisions towards their goals and a shared environment. The novelty of this approach is that it captures emergent behaviour and often counter-intuitive results by analyzing at the individual agent level (which is often highly heterogeneous). Bonabeau explains this modeling has been applied to many fields such as transportation, supermarket design, and stock markets.

Despite this broad applicability of ABM, it must be approached cautiously. I believe that there is a large gap between seeing a simulation of an emergent phenomenon and whether it can be validated as representative of reality. The accuracy of these simulations depends on the inputted parameters, which often must reflect difficult-to-quantify behaviours. An uncritical acceptance of ABM’s results can risk large sums of money, public trust, and lives.  

Furthermore, it is important to use ABM to its full potential. Users of this tool should not focus solely on running this model until they get a desired result. There is room for geographic analysis of unexpected emergent interactions to better explain conclusions. There is also a need for a deep understanding of the limited spatial analysis each agent is capable of, and how the agents’ perception of their spatial surroundings affects their behaviour.   

Bonabeau, Eric. “Agent-based Modeling: Methods and Techniques for Simulating Human Systems.” Proceedings of the National Academy of Sciences of the United States of America. 99.10 (2002): 7280-7287. Print.

– Madskiier_JWong