https://airwiki.elet.polimi.it/api.php?action=feedcontributions&user=MaurizioGarbarino&feedformat=atomAIRWiki - User contributions [en]2024-03-28T12:57:25ZUser contributionsMediaWiki 1.25.6https://airwiki.elet.polimi.it/index.php?title=User_talk:MaurizioGarbarino&diff=12827User talk:MaurizioGarbarino2011-01-09T21:40:59Z<p>MaurizioGarbarino: /* Questions */</p>
<hr />
<div>== Questions ==<br />
<br />
1.1) Is it possible to formalize a general model valid for a large number of video game genres?<br />
1.2) TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or...<br />
1.3) Which video games could be used for the experiment? Our previous work was based on TORCS. However, it might be interesting to demonstrate that the model of entertainment obtained is general enough to be successfully used<br />
with other video games as well.<br />
Towards this direction I'm trying to perform tests using data from two different experiments (our on TORCS and Yannakakis' one with Mazeball). I think some theoretical assumptions have to be done in order to correctly interpret the results. The answer to this question can be the main topic of an article. Testing whether the models obtained from a game can successfully predict the enjoyment on a second game (and vice-versa) could help to answer to these questions.<br />
<br />
<br />
2.1) Which aspect of a video game can be modified to best stimulate the player? How can these be modified at run-time to follow user physiological signals?<br />
Several issues have to be investigated such as the optimal time window for adaptation that can depend on the reactivity of the game controller and on how sensible is player response to the given change of status in the game.<br />
2.2) How to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
This questions are really interesting but I think that a specific experiment is needed to answer to every single question. I do not think it is possible to get useful information from the data we have collected during our experiment on TORCS. Thinking about the thesis work of the biomedical guys unfortunately we did not get any reliable results. Although we do have now the full working setup (the interaction between TORCS, the physiological capture device, and the enjoyment model estimator in Matlab) for dynamic adaptation and real time enjoyment estimation. If we think to the right, and rather specific, questions it will be possible to setup some ad-hoc experiments for each questions. As usual, it is fundamental to deeply thing about what we want address *before* starting to perform the experiment :).<br />
<br />
Regarding the question 2.2, I thing it is really complementary to the question 2.1. To create a game that successfully adapt its behavior, we have to know what are the parameters that can be evolved and how the player react to them. In my opinion, these issues are more "humanistic" oriented and probably it is possible to find some discussion about it in the literature. I am not really used to theoretical assumption (maybe I would rather say that I am more used to address problems that need to be "solved"). What I mean is that I do not have a clear picture on how to proceed to answer the 2.2 question.<br />
<br />
<br />
3.1) How does the adaptation by physiological signals compare to the adaptation by performance analysis? <br />
3.2) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
I think this can be "easily" addressed once I will be back and I can meet the student that is working on performance analysis. The comparison should be straight forward and we can try to build a model that keeps the best of the two worlds. Although it is fundamental to not mix the task recognition with the enjoyment recognition. For example in our experiment, it is easy to have a misleading result regarding the button pressing activity: during the ''challenging'' race (as opposite to the other 2 ''boring'' race situation) the player has to fight with the opponent to overtake it (i.e., more button are pressed faster!). This complete analysis of physiological and performance data can be the main topic of an article. Question 3.2 can be addressed also by integrating and comparing works on other experiment on adaptation in videogames in literature.<br />
<br />
<br />
4) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
Probably, to answer this question, a specific control experiment is needed. In alternative, to use the data from our TORCS experiment, it might be possible to train a model that classifies the task and compare it to the model that classifies the enjoyment. Then, we could compare the performance but I am not sure that this approach can really help to answer the question. The main issue here is that often the task and the preference overlap (I mean that players often prefer the challenging race) and therefore the 2 models result to be similar.<br />
<br />
<br />
------------------<br />
Questo è il riassunto di tutto quello che ci siamo detti prima della mia partenza.<br />
<br />
== Deadlines ==<br />
<br />
Qua di seguito ci sono le date importanti per le 2 conferenze da non mancare. Non ho trovato altri eventi interessanti a cui poter puntare. Se ne dovessero saltare fuori li aggiungiamo qua sotto!<br />
<br />
[http://cilab.sejong.ac.kr/cig2011/ CIG2011]<br />
<br />
January 31, 2011 Tutorial and special session proposal deadline<br />
<br />
'''March 15, 2011''' - Paper submission deadline<br />
<br />
August 31, 2011 - Conference starts<br />
<br />
<br />
<br />
[http://www.acii2011.org/ ACII2011]<br />
<br />
'''April 1''', 2011 - Regular papers due <br />
<br />
April 15, 2011 - Abstracts for demos due <br />
<br />
June 1, 2011 - Acceptance notification <br />
<br />
June 21, 2011 - Final versions due <br />
<br />
October 9, 2011 - Conference starts <br />
<br />
<br />
<br />
== Referenze ==<br />
<br />
Qua di seguito riporto le parti salienti delle mail che ci siamo scambiati, anche con Yannakakis giusto per referenza.<br />
<br />
In my PhD studies I have been investigating whether physiological<br />
signals can be used as means for video game adaptation.<br />
With the aim of keeping high the level of enjoyment of the player, the<br />
first step was building a model for enjoyment estimation from<br />
physiological signals.<br />
So far, we obtained promising results on the relationship between<br />
variation of physiological signals and variation of enjoyment during a<br />
TORCS game session.<br />
The next step is to study how to exploit the obtained model for the<br />
actual dynamic adaptation of it.<br />
During my staying at your Lab, I would like to focus on the analysis,<br />
design and development of this second step.<br />
Here are some key aspects that i would like to investigate:<br />
<br />
- Which aspect of a video game can be modified to best stimulate the<br />
player? How can these be modified at run-time to follow user<br />
physiological signals?<br />
Several issues have to be investigated such as the optimal time window<br />
for adaptation that can depend on the reactivity of the game<br />
controller and on how sensible is player response to the given change<br />
of status in the game.<br />
<br />
- Is it possible to formalize a general model valid for a large number<br />
of video game genres?<br />
<br />
- How does the adaptation by physiological signals compare to the<br />
adaptation by performance analysis?<br />
<br />
- Which video games could be used for the experiment?<br />
Our previous work was based on TORCS. However, it might be interesting<br />
to demonstrate that the model of entertainment obtained is general<br />
enough to be successfully used with other video games as well.<br />
<br />
---------------------------<br />
The research questions you pose are interesting indeed. There are three main directions we could investigate<br />
while you are here:<br />
1) the adaptation mechanism for TORCS (as you propose) and/or<br />
2) steps towards re-designing the models via dissimilar experimental protocols and alternative modeling techniques and/or<br />
3) investigations for the generalizability of the models.<br />
We can discuss the details of these directions while you are here of course.<br />
<br />
-----------------<br />
Dear all, I think all the three lines are perfectly ok. If I should express my *preference* I would rank those as 2, 3, and 1.<br />
<br />
It would be great if Maurizio could exploit the experience in game design to @itu and define how at theoretical and practical level issues of affective computing:<br />
<br />
- how to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
- what is the reasonable adaptation rate with this kind of feedback (i.e., how long does it takes to have a steady state responce)? Are we really "controlling" the subject to keep her in the flow? How?<br />
- TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or ...<br />
<br />
All of this might require or lead to the answers we all are aiming at:<br />
<br />
1) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
2) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
<br />
I know this broaden even more the project proposal, but having all the ingredients on the table might help to understand the recipe ;-)<br />
----------------------<br />
<br />
Quello che manca per dare sostanza al lavoro finale, secondo me, è la parte di adattamento, sia analisi sulle problematiche (tempistiche, possibili parametri su cui agire ecc...), sia dei risultati su un esperimento completo.<br />
------------------------<br />
<br />
<br />
Il primo obiettivo che abbiamo stabilito è l'integrazione dell'algoritmo che hanno presentato a CIG per la stima della funzione di preferenza. Al posto del metodo lineare che abbiamo usato noi, loro usano una rete neurale appresa tramite un algoritmo genetico. Fatto questo, proverò ad applicare il nostro metodo lineare ai loro dati per poi vedere come si comporta il modello appreso su un esperimento quando viene applicato ad un altro set di dati derivanti da un'altra tipologia di esperimento. <br />
<br />
<br />
Lo scorso meeting con Yannakakis abbiamo discusso sulla criticità della qualità della normalizzazione dei segnali dati in ingresso alla rete neurale nell'approcio neuroevolutionary. Infatti, dai primi test che avevo fatto, erano emersi due problemi: diverse esecuzioni dell'algoritmo di feature selection davano sottoinsiemi di features molto disomogenei e con performance che variavano dal 72% all'80%. E inoltre, alcune delle features selezionate erano sospette in quanto "ad occhio" non avremmo mai detto che sarebbero rientrate nella selezione, come ad esempio il delta tra il tempo del valore minimo e il tempo del valore massimo dell'heart rate o della temperatura che sulla carta non dovrebbero essere molto discriminanti.<br />
Altro problema, con il numero di features che abbiamo, una rete neurale complessa ci mette molto tempo per convergere ad una soluzione ottima. L'ultimo pensiero va al fatto che le performance variano molto a seconda del fold considerato (durante la crossvalidazione). Questo è sintomo del fatto che potrebbero esserci dei cluster di player per cui un approccio di player modeling potrebe portare a migliori performance. Però su quest'ultimo aspetto non credo di avere tempo a sufficienza per investigare...<br />
<br />
---------------------------------</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User_talk:MaurizioGarbarino&diff=12826User talk:MaurizioGarbarino2011-01-09T21:39:28Z<p>MaurizioGarbarino: /* Questions */</p>
<hr />
<div>== Questions ==<br />
<br />
1.1) Is it possible to formalize a general model valid for a large number of video game genres?<br />
1.2) TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or...<br />
1.3) Which video games could be used for the experiment? Our previous work was based on TORCS. However, it might be interesting to demonstrate that the model of entertainment obtained is general enough to be successfully used with other video games as well.<br />
Towards this direction I'm trying to perform tests using data from two different experiments (our on TORCS and Yannakakis' one with Mazeball). I think some theoretical assumptions have to be done in order to correctly interpret the results. The answer to this question can be the main topic of an article. Testing whether the models obtained from a game can successfully predict the enjoyment on a second game (and vice-versa) could help to answer to these questions.<br />
<br />
<br />
2.1) Which aspect of a video game can be modified to best stimulate the player? How can these be modified at run-time to follow user physiological signals?<br />
Several issues have to be investigated such as the optimal time window for adaptation that can depend on the reactivity of the game controller and on how sensible is player response to the given change of status in the game.<br />
2.2) How to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
This questions are really interesting but I think that a specific experiment is needed to answer to every single question. I do not think it is possible to get useful information from the data we have collected during our experiment on TORCS. Thinking about the thesis work of the biomedical guys unfortunately we did not get any reliable results. Although we do have now the full working setup (the interaction between TORCS, the physiological capture device, and the enjoyment model estimator in Matlab) for dynamic adaptation and real time enjoyment estimation. If we think to the right, and rather specific, questions it will be possible to setup some ad-hoc experiments for each questions. As usual, it is fundamental to deeply thing about what we want address *before* starting to perform the experiment :).<br />
<br />
Regarding the question 2.2, I thing it is really complementary to the question 2.1. To create a game that successfully adapt its behavior, we have to know what are the parameters that can be evolved and how the player react to them. In my opinion, these issues are more "humanistic" oriented and probably it is possible to find some discussion about it in the literature. I am not really used to theoretical assumption (maybe I would rather say that I am more used to address problems that need to be "solved"). What I mean is that I do not have a clear picture on how to proceed to answer the 2.2 question.<br />
<br />
<br />
3.1) How does the adaptation by physiological signals compare to the adaptation by performance analysis? <br />
3.2) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
I think this can be "easily" addressed once I will be back and I can meet the student that is working on performance analysis. The comparison should be straight forward and we can try to build a model that keeps the best of the two worlds. Although it is fundamental to not mix the task recognition with the enjoyment recognition. For example in our experiment, it is easy to have a misleading result regarding the button pressing activity: during the ''challenging'' race (as opposite to the other 2 ''boring'' race situation) the player has to fight with the opponent to overtake it (i.e., more button are pressed faster!). This complete analysis of physiological and performance data can be the main topic of an article. Question 3.2 can be addressed also by integrating and comparing works on other experiment on adaptation in videogames in literature.<br />
<br />
<br />
4) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
Probably, to answer this question, a specific control experiment is needed. In alternative, to use the data from our TORCS experiment, it might be possible to train a model that classifies the task and compare it to the model that classifies the enjoyment. Then, we could compare the performance but I am not sure that this approach can really help to answer the question. The main issue here is that often the task and the preference overlap (I mean that players often prefer the challenging race) and therefore the 2 models result to be similar.<br />
<br />
<br />
------------------<br />
Questo è il riassunto di tutto quello che ci siamo detti prima della mia partenza.<br />
<br />
<br />
<br />
== Deadlines ==<br />
<br />
Qua di seguito ci sono le date importanti per le 2 conferenze da non mancare. Non ho trovato altri eventi interessanti a cui poter puntare. Se ne dovessero saltare fuori li aggiungiamo qua sotto!<br />
<br />
[http://cilab.sejong.ac.kr/cig2011/ CIG2011]<br />
<br />
January 31, 2011 Tutorial and special session proposal deadline<br />
<br />
'''March 15, 2011''' - Paper submission deadline<br />
<br />
August 31, 2011 - Conference starts<br />
<br />
<br />
<br />
[http://www.acii2011.org/ ACII2011]<br />
<br />
'''April 1''', 2011 - Regular papers due <br />
<br />
April 15, 2011 - Abstracts for demos due <br />
<br />
June 1, 2011 - Acceptance notification <br />
<br />
June 21, 2011 - Final versions due <br />
<br />
October 9, 2011 - Conference starts <br />
<br />
<br />
<br />
== Referenze ==<br />
<br />
Qua di seguito riporto le parti salienti delle mail che ci siamo scambiati, anche con Yannakakis giusto per referenza.<br />
<br />
In my PhD studies I have been investigating whether physiological<br />
signals can be used as means for video game adaptation.<br />
With the aim of keeping high the level of enjoyment of the player, the<br />
first step was building a model for enjoyment estimation from<br />
physiological signals.<br />
So far, we obtained promising results on the relationship between<br />
variation of physiological signals and variation of enjoyment during a<br />
TORCS game session.<br />
The next step is to study how to exploit the obtained model for the<br />
actual dynamic adaptation of it.<br />
During my staying at your Lab, I would like to focus on the analysis,<br />
design and development of this second step.<br />
Here are some key aspects that i would like to investigate:<br />
<br />
- Which aspect of a video game can be modified to best stimulate the<br />
player? How can these be modified at run-time to follow user<br />
physiological signals?<br />
Several issues have to be investigated such as the optimal time window<br />
for adaptation that can depend on the reactivity of the game<br />
controller and on how sensible is player response to the given change<br />
of status in the game.<br />
<br />
- Is it possible to formalize a general model valid for a large number<br />
of video game genres?<br />
<br />
- How does the adaptation by physiological signals compare to the<br />
adaptation by performance analysis?<br />
<br />
- Which video games could be used for the experiment?<br />
Our previous work was based on TORCS. However, it might be interesting<br />
to demonstrate that the model of entertainment obtained is general<br />
enough to be successfully used with other video games as well.<br />
<br />
---------------------------<br />
The research questions you pose are interesting indeed. There are three main directions we could investigate<br />
while you are here:<br />
1) the adaptation mechanism for TORCS (as you propose) and/or<br />
2) steps towards re-designing the models via dissimilar experimental protocols and alternative modeling techniques and/or<br />
3) investigations for the generalizability of the models.<br />
We can discuss the details of these directions while you are here of course.<br />
<br />
-----------------<br />
Dear all, I think all the three lines are perfectly ok. If I should express my *preference* I would rank those as 2, 3, and 1.<br />
<br />
It would be great if Maurizio could exploit the experience in game design to @itu and define how at theoretical and practical level issues of affective computing:<br />
<br />
- how to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
- what is the reasonable adaptation rate with this kind of feedback (i.e., how long does it takes to have a steady state responce)? Are we really "controlling" the subject to keep her in the flow? How?<br />
- TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or ...<br />
<br />
All of this might require or lead to the answers we all are aiming at:<br />
<br />
1) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
2) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
<br />
I know this broaden even more the project proposal, but having all the ingredients on the table might help to understand the recipe ;-)<br />
----------------------<br />
<br />
Quello che manca per dare sostanza al lavoro finale, secondo me, è la parte di adattamento, sia analisi sulle problematiche (tempistiche, possibili parametri su cui agire ecc...), sia dei risultati su un esperimento completo.<br />
------------------------<br />
<br />
<br />
Il primo obiettivo che abbiamo stabilito è l'integrazione dell'algoritmo che hanno presentato a CIG per la stima della funzione di preferenza. Al posto del metodo lineare che abbiamo usato noi, loro usano una rete neurale appresa tramite un algoritmo genetico. Fatto questo, proverò ad applicare il nostro metodo lineare ai loro dati per poi vedere come si comporta il modello appreso su un esperimento quando viene applicato ad un altro set di dati derivanti da un'altra tipologia di esperimento. <br />
<br />
<br />
Lo scorso meeting con Yannakakis abbiamo discusso sulla criticità della qualità della normalizzazione dei segnali dati in ingresso alla rete neurale nell'approcio neuroevolutionary. Infatti, dai primi test che avevo fatto, erano emersi due problemi: diverse esecuzioni dell'algoritmo di feature selection davano sottoinsiemi di features molto disomogenei e con performance che variavano dal 72% all'80%. E inoltre, alcune delle features selezionate erano sospette in quanto "ad occhio" non avremmo mai detto che sarebbero rientrate nella selezione, come ad esempio il delta tra il tempo del valore minimo e il tempo del valore massimo dell'heart rate o della temperatura che sulla carta non dovrebbero essere molto discriminanti.<br />
Altro problema, con il numero di features che abbiamo, una rete neurale complessa ci mette molto tempo per convergere ad una soluzione ottima. L'ultimo pensiero va al fatto che le performance variano molto a seconda del fold considerato (durante la crossvalidazione). Questo è sintomo del fatto che potrebbero esserci dei cluster di player per cui un approccio di player modeling potrebe portare a migliori performance. Però su quest'ultimo aspetto non credo di avere tempo a sufficienza per investigare...<br />
<br />
---------------------------------</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User_talk:MaurizioGarbarino&diff=12825User talk:MaurizioGarbarino2011-01-09T21:07:32Z<p>MaurizioGarbarino: </p>
<hr />
<div>== Questions ==<br />
<br />
1.1) Is it possible to formalize a general model valid for a large number of video game genres?<br />
1.2) TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or...<br />
1.3) Which video games could be used for the experiment? Our previous work was based on TORCS. However, it might be interesting to demonstrate that the model of entertainment obtained is general enough to be successfully used with other video games as well.<br />
Towards this direction I'm trying to perform tests using data from two different experiments (our on TORCS and Yannakakis' one with Mazeball). I think some theoretical assumptions have to be done in order to correctly interpret the results. The answer to this question can be the main topic of an article. Testing whether the models obtained from a game can successfully predict the enjoyment on a second game (and vice-versa) could help to answer to these questions.<br />
<br />
<br />
2.1) Which aspect of a video game can be modified to best stimulate the player? How can these be modified at run-time to follow user physiological signals?<br />
Several issues have to be investigated such as the optimal time window for adaptation that can depend on the reactivity of the game controller and on how sensible is player response to the given change of status in the game.<br />
2.2) How to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
This questions are really interesting but I think that a specific experiment is needed to answer to every single question. I do not think it is possible to get useful information from the data we have collected during our experiment on TORCS. Thinking about the thesis work of the biomedical guys unfortunately we did not get any reliable results. Although we do have now the full working setup (the interaction between TORCS, the physiological capture device, and the enjoyment model estimator in Matlab) for dynamic adaptation and real time enjoyment estimation. If we think to the right, and rather specific, questions it will be possible to setup some ad-hoc experiments for each questions. As usual, it is fundamental to deeply thing about what we want address *before* starting to perform the experiment :).<br />
<br />
Regarding the question 2.2, I thing it is really complementary to the question 2.1. To create a game that successfully adapt its behavior, we have to know what are the parameters that can be evolved and how the player react to them. In my opinion, these issues are more "humanistic" oriented and probably it is possible to find some discussion about it in the literature. I am not really used to theoretical assumption (maybe I would rather say that I am more used to address problems that need to be "solved"). What I mean is that I do not have a clear picture on how to proceed to answer the 2.2 question.<br />
<br />
<br />
3.1) How does the adaptation by physiological signals compare to the adaptation by performance analysis? <br />
3.2) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
I think this can be "easily" addressed once I will be back and I can meet the student that is working on performance analysis. The comparison should be straight forward and we can try to build a model that keeps the best of the two worlds. Although it is fundamental to not mix the task recognition with the enjoyment recognition. For example in our experiment, it is easy to have a misleading result regarding the button pressing activity: during the ''challenging'' race (as opposite to the other 2 ''boring'' race situation) the player has to fight with the opponent to overtake it (i.e., more button are pressed faster!). This complete analysis of physiological and performance data can be the main topic of an article. Question 3.2 can be addressed also by integrating and comparing works on other experiment on adaptation in videogames in literature.<br />
<br />
<br />
4) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
Probably to answer this question a specific control experiment is needed. In alternative, Or, we have to think about a re <br />
<br />
<br />
<br />
<br />
<br />
<br />
Qua di seguito riporto le parti salienti delle mail che ci siamo scambiati, anche con Yannakakis giusto per referenza.<br />
----------<br />
<br />
In my PhD studies I have been investigating whether physiological<br />
signals can be used as means for video game adaptation.<br />
With the aim of keeping high the level of enjoyment of the player, the<br />
first step was building a model for enjoyment estimation from<br />
physiological signals.<br />
So far, we obtained promising results on the relationship between<br />
variation of physiological signals and variation of enjoyment during a<br />
TORCS game session.<br />
The next step is to study how to exploit the obtained model for the<br />
actual dynamic adaptation of it.<br />
During my staying at your Lab, I would like to focus on the analysis,<br />
design and development of this second step.<br />
Here are some key aspects that i would like to investigate:<br />
<br />
- Which aspect of a video game can be modified to best stimulate the<br />
player? How can these be modified at run-time to follow user<br />
physiological signals?<br />
Several issues have to be investigated such as the optimal time window<br />
for adaptation that can depend on the reactivity of the game<br />
controller and on how sensible is player response to the given change<br />
of status in the game.<br />
<br />
- Is it possible to formalize a general model valid for a large number<br />
of video game genres?<br />
<br />
- How does the adaptation by physiological signals compare to the<br />
adaptation by performance analysis?<br />
<br />
- Which video games could be used for the experiment?<br />
Our previous work was based on TORCS. However, it might be interesting<br />
to demonstrate that the model of entertainment obtained is general<br />
enough to be successfully used with other video games as well.<br />
<br />
<br />
---------------------------<br />
<br />
The research questions you pose are interesting indeed. There are three main directions we could investigate<br />
while you are here:<br />
1) the adaptation mechanism for TORCS (as you propose) and/or<br />
2) steps towards re-designing the models via dissimilar experimental protocols and alternative modeling techniques and/or<br />
3) investigations for the generalizability of the models.<br />
We can discuss the details of these directions while you are here of course.<br />
<br />
-----------------<br />
<br />
Dear all, I think all the three lines are perfectly ok. If I should express my *preference* I would rank those as 2, 3, and 1.<br />
<br />
It would be great if Maurizio could exploit the experience in game design to @itu and define how at theoretical and practical level issues of affective computing:<br />
<br />
- how to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
- what is the reasonable adaptation rate with this kind of feedback (i.e., how long does it takes to have a steady state responce)? Are we really "controlling" the subject to keep her in the flow? How?<br />
- TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or ...<br />
<br />
All of this might require or lead to the answers we all are aiming at:<br />
<br />
1) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
2) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
<br />
I know this broaden even more the project proposal, but having all the ingredients on the table might help to understand the recipe ;-)<br />
----------------------<br />
<br />
Quello che manca per dare sostanza al lavoro finale, secondo me, è la parte di adattamento, sia analisi sulle problematiche (tempistiche, possibili parametri su cui agire ecc...), sia dei risultati su un esperimento completo.<br />
<br />
------------------------<br />
<br />
<br />
Il primo obiettivo che abbiamo stabilito è l'integrazione dell'algoritmo che hanno presentato a CIG per la stima della funzione di preferenza. Al posto del metodo lineare che abbiamo usato noi, loro usano una rete neurale appresa tramite un algoritmo genetico. Fatto questo, proverò ad applicare il nostro metodo lineare ai loro dati per poi vedere come si comporta il modello appreso su un esperimento quando viene applicato ad un altro set di dati derivanti da un'altra tipologia di esperimento. <br />
<br />
<br />
Lo scorso meeting con Yannakakis abbiamo discusso sulla criticità della qualità della normalizzazione dei segnali dati in ingresso alla rete neurale nell'approcio neuroevolutionary. Infatti, dai primi test che avevo fatto, erano emersi due problemi: diverse esecuzioni dell'algoritmo di feature selection davano sottoinsiemi di features molto disomogenei e con performance che variavano dal 72% all'80%. E inoltre, alcune delle features selezionate erano sospette in quanto "ad occhio" non avremmo mai detto che sarebbero rientrate nella selezione, come ad esempio il delta tra il tempo del valore minimo e il tempo del valore massimo dell'heart rate o della temperatura che sulla carta non dovrebbero essere molto discriminanti.<br />
Altro problema, con il numero di features che abbiamo, una rete neurale complessa ci mette molto tempo per convergere ad una soluzione ottima. L'ultimo pensiero va al fatto che le performance variano molto a seconda del fold considerato (durante la crossvalidazione). Questo è sintomo del fatto che potrebbero esserci dei cluster di player per cui un approccio di player modeling potrebe portare a migliori performance. Però su quest'ultimo aspetto non credo di avere tempo a sufficienza per investigare...<br />
<br />
---------------------------------</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User_talk:MaurizioGarbarino&diff=12824User talk:MaurizioGarbarino2011-01-08T12:39:28Z<p>MaurizioGarbarino: New page: ---DRAFT--- In my PhD studies I have been investigating whether physiological signals can be used as means for video game adaptation. With the aim of keeping high the level of enjoyment o...</p>
<hr />
<div>---DRAFT---<br />
<br />
In my PhD studies I have been investigating whether physiological<br />
signals can be used as means for video game adaptation.<br />
With the aim of keeping high the level of enjoyment of the player, the<br />
first step was building a model for enjoyment estimation from<br />
physiological signals.<br />
So far, we obtained promising results on the relationship between<br />
variation of physiological signals and variation of enjoyment during a<br />
TORCS game session.<br />
The next step is to study how to exploit the obtained model for the<br />
actual dynamic adaptation of it.<br />
During my staying at your Lab, I would like to focus on the analysis,<br />
design and development of this second step.<br />
Here are some key aspects that i would like to investigate:<br />
<br />
- Which aspect of a video game can be modified to best stimulate the<br />
player? How can these be modified at run-time to follow user<br />
physiological signals?<br />
Several issues have to be investigated such as the optimal time window<br />
for adaptation that can depend on the reactivity of the game<br />
controller and on how sensible is player response to the given change<br />
of status in the game.<br />
<br />
- Is it possible to formalize a general model valid for a large number<br />
of video game genres?<br />
<br />
- How does the adaptation by physiological signals compare to the<br />
adaptation by performance analysis?<br />
<br />
- Which video games could be used for the experiment?<br />
Our previous work was based on TORCS. However, it might be interesting<br />
to demonstrate that the model of entertainment obtained is general<br />
enough to be successfully used with other video games as well.<br />
<br />
<br />
---------------------------<br />
<br />
<br />
<br />
The research questions you pose are interesting indeed. There are three main directions we could investigate<br />
while you are here: 1) the adaptation mechanism for TORCS (as you propose) and/or 2) steps towards re-designing the models via<br />
dissimilar experimental protocols and alternative modeling techniques and/or 3) investigations for the generalizability of the models.<br />
We can discuss the details of these directions while you are here of course.<br />
<br />
-----------------<br />
<br />
<br />
Dear all, I think all the three lines are perfectly ok. If I should express my *preference* I would rank those as 2, 3, and 1.<br />
<br />
It would be great if Maurizio could exploit the experience in game design to @itu and define how at theoretical and practical level issues of affective computing:<br />
<br />
- how to design a game that could be adapted using biofeedback (meaning affective computing, eye tracking, image processing, whatever)?<br />
- what is the reasonable adaptation rate with this kind of feedback (i.e., how long does it takes to have a steady state responce)? Are we really "controlling" the subject to keep her in the flow? How?<br />
- TORCS is one example, he could proceed with that, but it would be great to define things more general and figure out if and how the protocol and the adaptivity could be verified in other games such as MsPacman or Mario or ...<br />
<br />
All of this might require or lead to the answers we all are aiming at:<br />
<br />
1) affective computing is really worth or performance analysis could be enough for adaptive gaming?<br />
2) are we really measuring the user fun/engagement or just the stimuli? I mean, the heart rate is increased because of engagement or just because different movements are requested to the user?<br />
<br />
I know this broaden even more the project proposal, but having all the ingredients on the table might help to understand the recipe ;-)<br />
----------------------<br />
<br />
Quello che manca per dare sostanza al lavoro finale, secondo me, è la parte di adattamento, sia analisi sulle problematiche (tempistiche, possibili parametri su cui agire ecc...), sia dei risultati su un esperimento completo.<br />
<br />
------------------------<br />
<br />
Colgo l'occasione per riassumerti come stanno andando le cose. L'ultima settimana ahimé l'ho passata a lavorare sulla relazione per il minore e sulla presentazione per l'ultimo esame (Microcontrollori & co. di Zappa).<br />
Lo scorso meeting con Yannakakis abbiamo discusso sulla criticità della qualità della normalizzazione dei segnali dati in ingresso alla rete neurale nell'approcio neuroevolutionary. Infatti, dai primi test che avevo fatto, erano emersi due problemi: diverse esecuzioni dell'algoritmo di feature selection davano sottoinsiemi di features molto disomogenei e con performance che variavano dal 72% all'80%. E inoltre, alcune delle features selezionate erano sospette in quanto "ad occhio" non avremmo mai detto che sarebbero rientrate nella selezione, come ad esempio il delta tra il tempo del valore minimo e il tempo del valore massimo dell'heart rate o della temperatura che sulla carta non dovrebbero essere molto discriminanti.<br />
Altro problema, con il numero di features che abbiamo, una rete neurale complessa ci mette molto tempo per convergere ad una soluzione ottima. L'ultimo pensiero va al fatto che le performance variano molto a seconda del fold considerato (durante la crossvalidazione). Questo è sintomo del fatto che potrebbero esserci dei cluster di player per cui un approccio di player modeling potrebe portare a migliori performance. Però su quest'ultimo aspetto non credo di avere tempo a sufficienza per investigare...<br />
<br />
---------------------------------</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=Affective_VideoGames&diff=11628Affective VideoGames2010-05-07T15:15:53Z<p>MaurizioGarbarino: </p>
<hr />
<div>{{Project<br />
|title=Affective VideoGames<br />
|coordinator=SimoneTognetti;MaurizioGarbarino;AndreaBonarini;MatteoMatteucci<br />
|tutor=SimoneTognetti; MaurizioGarbarino; AndreaBonarini; MatteoMatteucci;<br />
|students=LucaPerego; AntonellaBelfatto; BarbaraBruno;GianmarcoZaccaria; <br />
|resarea=Affective Computing<br />
|restopic=Affective Computing And BioSignals<br />
|start=2008/11/14<br />
|end=2010/12/23<br />
|status=Active<br />
|level=Ms<br />
}}<br />
=== Project description ===<br />
The aim of this projects is to develop a software that can adapt a behaviour of an interactive videogame (a car game like TORCS or a similar parametrizable game) according to an evaluation of his/her emotional state. We will start by analizing biological signals to evaluate the emotional state.<br />
The goal is to maximize the engagement of the player, making her/him staying as long as possible in the "flow" state , by adapting the parameters of the videogame to her/his real emotional state.<br />
<br />
The first part of the project consisted of the analysis of a set of VideoGames. In particular three different available open source games have been studied in order to receive feedbacks from the player (in terms of excitement), finding out which one is the most suitable for the goal of the project: the one potentially affecting most the player's emotions. <br />
The games that have been analyzed for this first stage of the project have been the following:<br />
<br />
- a puzzle/action 3d game called Beaver Valley;<br />
<br />
- TORCS (driving simulator);<br />
<br />
- the music video game called Frets on Fire.<br />
<br />
TORCS has been selected and experiments with it have been done in October 2009, leading to the definition of a methodology to face experiments and some preliminary results. <br />
<br />
Here is an example of this experiment. The player is sitting in front of a desktop computer equipped with a keyboard. During the game the biological signal of the subject are acquired by sensors on the fingers and the [[Biofeedback_and_neurofeedback_systems|ProComp Infiniti]] system. <br />
<br />
{{#ev:youtube|XD8C19BUVUg}}<br />
<br />
*[http://www.youtube.com/watch?v=XD8C19BUVUg External link]<br />
<br />
Following this methodology, in March 2010 a new set of experiments have been performed with 80 volunteers. In these experiments, images from two cameras have also been recorded to allow the [[Gestures in Videogames|analysis of gestures and facial expressions]]. [[user:LucaPerego|Luca Perego]] is working on these data and on biophysical data to establish relationships between the two.<br />
<br />
From April 2010 [[user:BarbaraBruno|Barbara Bruno]] and [[user:AntonellaBelfatto|Antonella Belfatto]] are working on collecting new data and [[Videogame adaptation|adapting the videogame behavior]] to the detected user preferences.<br />
<br />
<br />
<br />
More details on the project are available under the discussion tab, accessible by involved airwiki users.<br />
<br />
===== Students that worked on the project in the past =====<br />
<br />
*[[User:AndreaCampana | Andrea Campana]] (Project work - Sep 2009)<br />
<br />
*[[User:AndreaTommasoBonanno|Andrea Tommaso Bonanno]] (MS Thesis - Dec 2009)<br />
<br />
====Laboratory work and risk analysis ====<br />
<br />
Laboratory work for this project will be mainly performed at AIRLab/DEI. The only risks are related to the use of computers and data acquisition devices</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=Videogame_adaptation&diff=11627Videogame adaptation2010-05-07T15:15:21Z<p>MaurizioGarbarino: </p>
<hr />
<div>{{Project<br />
|title=Videogame adaptation<br />
|short_descr=Adaptation of TORCS to match the user preferences detected from physiological signals<br />
|coordinator=AndreaBonarini<br />
|tutor=SimoneTognetti; MaurizioGarbarino;<br />
|students=AntonellaBelfatto; BarbaraBruno; GianmarcoZaccaria; <br />
|resarea=Affective Computing<br />
|restopic=Affective Computing And BioSignals;<br />
|start=2010/04/01<br />
|end=2010/09/30<br />
|status=Active<br />
|level=Ms<br />
|type=Course<br />
}}<br />
This project, belonging to the [[Affective VideoGames]] research line, is aimed at closing the affective adaptation loop on TORCS. Physiological data collected from users playing TORCS are used to model their preferences and this model is in turn used to adapt the game so to have it matching the user preferences as felt by the user.<br />
<br />
The project is done as part of a course in the Biomedical Engineering Master Track held by prof. [http://www.biomed.polimi.it/BioIntro/personale/docente/bianchi.htm Anna Maria Bianchi]</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User:GianmarcoZaccaria&diff=11626User:GianmarcoZaccaria2010-05-07T15:14:04Z<p>MaurizioGarbarino: </p>
<hr />
<div>{{Student<br />
|category=Student<br />
|firstname=Gianmarco<br />
|lastname=Zaccaria<br />
|email=gianmarco.zaccaria@gmail.com<br />
|projectpage=Affective_VideoGames<br />
|advisor=SimoneTognetti; MaurizioGarbarino;<br />
|status=active<br />
}}</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User:GianmarcoZaccaria&diff=11625User:GianmarcoZaccaria2010-05-07T15:11:56Z<p>MaurizioGarbarino: New page: {{Student |category=Student |firstname=Gianmarco |lastname=Zaccaria |email=gianmarco.zaccaria@gmail.com |advisor=SimoneTognetti; MaurizioGarbarino; |status=active }}</p>
<hr />
<div>{{Student<br />
|category=Student<br />
|firstname=Gianmarco<br />
|lastname=Zaccaria<br />
|email=gianmarco.zaccaria@gmail.com<br />
|advisor=SimoneTognetti; MaurizioGarbarino;<br />
|status=active<br />
}}</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User:MaurizioGarbarino&diff=5089User:MaurizioGarbarino2009-02-02T07:48:57Z<p>MaurizioGarbarino: </p>
<hr />
<div>{{UserProf<br />
|firstname=Maurizio<br />
|lastname=Garbarino<br />
|email=garbarino(at)elet(dot)polimi(dot)it<br />
|researchareas=Affective Computing<br />
* [[Affective Computing]]<br />
}}<br />
<br />
== Personal info! ==<br />
I was born in Torino in 1983, I got my bachelor's degree in Computer Science Engineering at the Politecnico di Torino in 2005.<br />
Then, i joined the double degree program and i spent about 2 years in Stockholm where i got my master's degree in Computer Science Engineering at [http://en.wikipedia.org/wiki/Royal_Institute_of_Technology Royal Institute of Technology] in 2008<br />
<br />
I am currently a PHD student at Politecnico di Milano working on "Affective computing". Our research group is experimenting a way to to introduce some affective process into totally cognitive intelligent systems. <br />
<br />
<br />
* [http://www.dei.polimi.it/people/garbarino Pagina DEI]</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User:MaurizioGarbarino&diff=5088User:MaurizioGarbarino2009-02-02T07:19:42Z<p>MaurizioGarbarino: </p>
<hr />
<div>{{UserProf<br />
|firstname=Maurizio<br />
|lastname=Garbarino<br />
|email=garbarino(at)elet(dot)polimi(dot)it<br />
|researchareas=Affective Computing<br />
* [[Affective Computing]]<br />
}}<br />
<br />
<br />
<br />
* [http://www.dei.polimi.it/people/garbarino Pagina DEI]</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User:MaurizioGarbarino&diff=5087User:MaurizioGarbarino2009-02-02T07:18:21Z<p>MaurizioGarbarino: </p>
<hr />
<div>{{SMWUser<br />
|firstname=Maurizio<br />
|lastname=Garbarino<br />
|email=garbarino(at)elet(dot)polimi(dot)it<br />
|advisor=AndreaBonarini<br />
|researchareas=Affective Computing<br />
* [[Affective Computing]]<br />
|photo=<br />
}}<br />
<br />
<br />
<br />
* [http://www.dei.polimi.it/people/garbarino Pagina DEI]<br />
<br />
<br />
{{#ask: [[hasAdvisor::User:MaurizioGarbarino]] <br />
| format=list<br />
| default=nothing found<br />
}}</div>MaurizioGarbarinohttps://airwiki.elet.polimi.it/index.php?title=User:MaurizioGarbarino&diff=5086User:MaurizioGarbarino2009-02-02T07:17:34Z<p>MaurizioGarbarino: New page: {{SMWUser |firstname=Maurizio |lastname=Garbarino |email=garbarino(at)elet(dot)polimi(dot)it |advisor=AndreaBonarini |projectpage= |photo= }} * [http://www.dei.polimi.it/people/garbarin...</p>
<hr />
<div>{{SMWUser<br />
|firstname=Maurizio<br />
|lastname=Garbarino<br />
|email=garbarino(at)elet(dot)polimi(dot)it<br />
|advisor=AndreaBonarini<br />
|projectpage=<br />
|photo=<br />
}}<br />
<br />
<br />
<br />
* [http://www.dei.polimi.it/people/garbarino Pagina DEI]<br />
<br />
<br />
{{#ask: [[hasAdvisor::User:MaurizioGarbarino]] <br />
| format=list<br />
| default=nothing found<br />
}}</div>MaurizioGarbarino