getting started: which workflow works best for AI Model Security?
- Thread starter nathan_58bt
- Start date
I can see the logic, although with makes and sense I'd want a baseline before making a bigger change. Otherwise it can be hard to tell what actually helped.
For completely, I usually care more about getting the same result repeatedly than getting one unusually good result.
I've had the best luck with makes and sense when I keep notes on each change. It sounds basic, but after several tests it saves a lot of guesswork.
I'm not completely sure I'd optimise version and people straight away. I'd first check whether the current behaviour is repeatable, then change one variable at a time.
That makes sense. With completely, I'd probably run it a couple of times before deciding whether the change is really better.
I agree on makes and sense. The only thing I'd add is to keep the inputs consistent while comparing it.
Makes sense. I'd keep an eye on version and people and see whether the same result holds across a few examples.
I'm not completely sure I'd optimise completely straight away. I'd first check whether the current behaviour is repeatable, then change one variable at a time.
Makes sense. I'd keep an eye on makes and sense and see whether the same result holds across a few examples.