Methods Week 10 - The future of research machine learning and AI (1/2)

what is machine learning?

application of AI that allows a system to automatically learn and improve from experience

deep learning

trying to emulate how humans work (via neuroscience neuron pathways)

what is AI?

Ability of a machine to imitate intelligent human behaviour

supervised learning

telling the machine learning thing what something is at face value.

ex. showing the machine a ball and then telling it t is a ball

“Supervised learning is where you have input variables (X) and an output variable (Y), and you use an algorithm to learn the mapping function from the input to the output.”

support vector machines (SMV)

training a software to separate data based on showing it different orientations and pictures of one image and then putting into a category. if it hasn’t been shown enough images to identify it, then it is not 100% accurately putting it in a certain category 

does machine learning use regression?

no, categorical data

if you tell someone their height they can put u in a weight category 

neural networks

start off with very low levels of information and then can make new understandings with each level of criteria

unsupervised learning

you only have input data (X) and no corresponding output variables

• factor analysis is a type of unsupervised learning

generative adversarial networks

the generator will be told why what they produces is real or fake bc of X and Y and then will put X and Y back into it’s model to produce another image and continues until it cannot distinguish between real or fake faces

how did chatgpt work

step 1-2 is supervised step 3-4 is unsupervised