AI and Computer Learning
A brief synopsis of this weeks lecture would be that it explored how technology may be used to assist us in daily life or even replace humans doing basic functions. Machines are being designed that gathers data and learns via this.This does come with its own risks though. Letting a machine know the details about our lives and movements opens it up to being hacked. This would mean sensitve details would could be stolen. It could also be dangerous in a sense that we could be easily tracked.
The idea of constantly being watched by something is the same concept of George Owells novel "1984" released in 1949, long before technology was actually capible of doing such a thing. This dystopian idea that the government will impose excessive security survalance on people in order to make sure they keep in line isnt far off reality in 2019. Facial recognition programs are in place that scan digital images to find someones likeness on a database.
An example of machine learning in practice is Hello Barbie. "Hello Barbie responds realistically to a child by using natural language processing, machine learning and advanced analytics to parse what the child says and respond accordingly. The doll has a microphone concealed in her necklace that records what anyone says to her and transmits it to the servers at ToyTalk. There, the recording is parsed, analysed, and the correct response selected and sent back for Barbie to deliver — all in under a second." (Bernard Marr & Co, 2019). The doll learns certain things about the user, remembering their favorite musician or favorite colour. This can be brought up weeks later in conversation. The technology also uses real voice lines which have to be recorded frequently based on conversations between dolls and their users. This can be used later on and makes sure the responses from the doll are kept current.
Machine learning is a complex subject and the possibilities are really endless. We are seeing the introduction to the technology in our everyday lives in things like driverless cars, amazon echoes and in our mobile phones.
The idea of constantly being watched by something is the same concept of George Owells novel "1984" released in 1949, long before technology was actually capible of doing such a thing. This dystopian idea that the government will impose excessive security survalance on people in order to make sure they keep in line isnt far off reality in 2019. Facial recognition programs are in place that scan digital images to find someones likeness on a database.
An example of machine learning in practice is Hello Barbie. "Hello Barbie responds realistically to a child by using natural language processing, machine learning and advanced analytics to parse what the child says and respond accordingly. The doll has a microphone concealed in her necklace that records what anyone says to her and transmits it to the servers at ToyTalk. There, the recording is parsed, analysed, and the correct response selected and sent back for Barbie to deliver — all in under a second." (Bernard Marr & Co, 2019). The doll learns certain things about the user, remembering their favorite musician or favorite colour. This can be brought up weeks later in conversation. The technology also uses real voice lines which have to be recorded frequently based on conversations between dolls and their users. This can be used later on and makes sure the responses from the doll are kept current.
Machine learning is a complex subject and the possibilities are really endless. We are seeing the introduction to the technology in our everyday lives in things like driverless cars, amazon echoes and in our mobile phones.
Comments
Post a Comment