@mathyawp.bsky.social and @stevenstrogatz.com both lending their voices to the AI conversation. Excellent.

Show Post

Here’s a simple demo of a 2D linear classifier in @desmos.com, using a pseudo-inverse matrix. I used to need NumPy for this. Now, with the new matrix features, it works in Desmos. Much better for being able to drag the points around! www.desmos.com/calculator/6... #iTeachMath

Show Post

Stewart? Larson? Sullivan? I feel like this some archaic AP definition that has gone counter to logic and all mainstream texts are in on it. Or are we sticklers for 2 sides limits for globals extrema? Either way, yeah, doesn't logically track.

Show Post

I'll just note that in APEX Calculus, we count endpoints as local extrema. And why not? The interval [-1,-0.9) is open in the subset topology

Show Post

I've always taught endpoints as potential local extrema. This fits the topology definition: f has is local max at c if there exists a neighborhood where f(c)≥f(x) for all x in that neighborhood. Neighborhoods at endpoints of closed intervals contain half open intervals.

Show Post