Friday, March 24, 2017

Topic: Psychophysics meets Machine Learning: Is machine classification all there is?


Speaker: Mick Smythwood


Abstract: Psychophysics, the branch of psychology that deals with the relationships between physical stimuli and mental phenomena, has benefited from machine learning techniques such as Support Vector Machines for item classification. Subjective measures such as ratings or rankings, as well as objective measures like eye-tracking or EEG readings, are used in psychophysics. Psychologists use traditional statistical methods such as PCA, maximum likelihood estimation, and equation modeling to analyze these types of measurements; however, it is the intermixing of traditional methods with machine learning techniques that has been found to have the most potential. In this talk we will discuss these methods and how they are being used today in the area of Experimental Aesthetics.