Earlier this week I checked and saw I wasn't enrolled even though I thought I was. I was a bit relieved. Yesterday I found out my current fasting AC was 245. I knew it was high but not that high. I had a filling fall out a week before Christmas and set up an appointment to have the tooth removed. It turned out it was just an exam. The surgeon wasn't going to remove the tooth anyway but after the exam he said he didn't want to remove it until I got my blood sugar under control. I was sort of hoping to not know and let nature take its course but I don't want to let the shell of a tooth just hang out because it's painful. I have trouble dealing with lots of stress and dealing with this is stressful.
Any way....
I'm enrolled and don't know why it didn't appear on my Coursera dashboard earlier this week.
The first week was pretty simple. I'll go ahead and go through the homework because my MatLab isn't that good. I've learned some new things in the lectures but he also covered stuff I learned in "Brains, Neurons, and Synapses". I'm glad I'm able to correlate what I learned in that course with stuff I get from this course. Unfortunately it looks like the dendrite model will be collapsed into a point neuron even though some of the delay caused by the length of the axon may be handled. I'm glad our model won't be of a point neuron with value representing firing rate as one of the suggested books has described its model. I bought both books several years ago but have misplaced the "Theoretical Neuroscience" text book when I moved into my current house a year and a half ago. It's a bit pricy on Amazon so will go without it now. I bought a copy of the other book again and was working through it, lacking enthusiasm because of the model. I'll hope for the best.
One of the major differences between point neurons and cable theory is that integration happens in the dendrites in cable theory. The ion flows in and out of the branches equal zero. The movement is generally towards the soma because the size of the branch is bigger closer to the soma and the soma itself is a big collection pool. I don't know, maybe there are action potentials at work all along the dendrite not just as the axonal hillock. The rule is suspend disbelief.
It's hard to think about the Hinton classifier, a Restricted Boltzmann Machine, while learning about more natural the models of the brain. I hope I can handle it. There are places where functionality takes precedence over accuracy.
No comments:
Post a Comment