Abstract
In this chapter, a partial solution to the grand challenge posed in the previous chapter is presented. Again, all the noological processing devices that were developed earlier are brought to bear on the problem, notably those from Chaps. 6 and 7. This further affirms that the devices developed in this book have sufficient generality and are applicable to a wide range of scenarios. It is found, in the process of addressing the SAS micro-environment benchmark, that a type of personality trait, neuroticism, has to be introduced to characterize the agent involved and this personality characterization influences the outcome of the problem solving process.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This can be inferred from just reading off the OUTCOME portion of the script or a mental simulation can be carried out through the use of the ACTIONS portion of the script. If the script is matched to the environment with high certainly, there is no need to carry out the mental simulation . However, similar to the situation as discussed in connection with Figs. 6.9 and 6.10 in Chap. 6, sometimes only a closest matched script is retrieved and there may be things in the environment that may affect the expected outcome of the script – a good example would be an obstacle between the Agent and the Projectile much like in the SMGO problem of Chap. 6. In that case, a simulation will reveal that the Projectile cannot reach the Agent so no future Pain event would ensue.
- 2.
There are two possibilities here. One possibility is that the match is not exact, given the EMAWAG-SCRIPT as it is represented in Fig. 6.19, and a moderate desperation in solving the current problem would result in the system accepting a certain non-perfect degree of match and the EMAWAG-SCRIPT is tried out. Another possibility is that the EMAWAG-SCRIPT could have earlier been generalized in terms of the shape and color of the Wall and the Object(X), and so the match between the current scenario and the SCENARIO in the EMAWAG-SCRIPT is exact. In fact, if the match is not exact, after the current process in which this best match is used to solve the current problem, and if the actions succeed in achieving the Not-Contact condition, then the EMAWAG-SCRIPT of Fig. 6.19 can be generalized immediately to dictate arbitrary shape and color for the two entities involved in the SCENARIO.
- 3.
This strategy could also be a result of some learning process but here we assume it is built-in.
- 4.
If the full process of the script of Fig. 9.3 is repeatedly used in the simulation of Fig. 9.9, it would be computationally expensive. However, as mentioned earlier, only the OUTCOME portion of the script needs to be checked for the consequence/result of the script’s application, therefore full simulation is not needed. Also, the Agent can generalize from the results of a few instances of similar simulation results to conclude that the repeated simulation of Fig. 9.9 would give rise to the same results.
- 5.
Suppose Projectiles are like bullets, there may be new technologies in the future that allow the bullets’ trajectories to be curved or programmed in any way the shooter wishes.
References
Ho, S.-B. (2013). A grand challenge for computational intelligence – A micro-environment benchmark for adaptive autonomous agents. In Proceedings of the IEEE symposium series on computational intelligence on intelligent agents (pp. 44–53) Singapore. Piscataway: IEEE Press.
Sagiv, L., Schwartz, S. H., & Knafo, A. (2002). The big five personality factors and personal values. Personality and Social Psychology Bulletin, 28, 789–801.
Author information
Authors and Affiliations
Problems
Problems
-
1.
Solve the rest of the SAS micro-environment as defined in Chap. 8
-
2.
At the end of Sect. 9.3 we mentioned an alternative solution to that of Fig. 9.17 in which the Agent “moves aside” for the Projectile to pass by, and that is to instead swing itself around some pivoting points to likewise avoid the Projectile. Devise a solution such as this using similar devices as we have discussed.
-
3.
Recently there has been attempts at using reinforcement learning to learn to play a number of “Atari games” such as Space Invaders (deepmind.com). The shooting of “bullets” between the player and the “space invaders,” and the player having to learn to move the object under her control to avoid the bullets has some similarity to the “projectile and agent” problem tackled in this chapter. Formulate an effective causal learning framework along the line of this book to learn to play games such as Space Invaders as well as other similar games.
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Ho, SB. (2016). Affect Driven Noological Processes. In: Principles of Noology. Socio-Affective Computing, vol 3. Springer, Cham. https://doi.org/10.1007/978-3-319-32113-4_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-32113-4_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-32111-0
Online ISBN: 978-3-319-32113-4
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)