1) Much more practice at teaching maths and physics, and it's much fresher in my mind, so I reckon you'll get higher quality out of me that way.
2) I expect more demand for maths than physics
3) A lot of physics and machine learning is maths anyway; particularly with machine learning, the bit of it I'm good at is more mathematical (statistics, bayesian inference, etc). When I took my MSc, the coding exercises were always harder for me (lack of background and formal training) but some of the mathematics was stuff that I'd actually been teaching for years. I seriously doubt I can teach anyone here much about implementing, well, anything, but if you want to get a good idea of probability theory and Bayesian inference, I'm more useful.
For clarity, though: I'm offering anything you request that I also think I can fulfil. Edited original post to make it a bit clearer.
2) I expect more demand for maths than physics
3) A lot of physics and machine learning is maths anyway; particularly with machine learning, the bit of it I'm good at is more mathematical (statistics, bayesian inference, etc). When I took my MSc, the coding exercises were always harder for me (lack of background and formal training) but some of the mathematics was stuff that I'd actually been teaching for years. I seriously doubt I can teach anyone here much about implementing, well, anything, but if you want to get a good idea of probability theory and Bayesian inference, I'm more useful.
For clarity, though: I'm offering anything you request that I also think I can fulfil. Edited original post to make it a bit clearer.