Speaker unverified: belief

Transcript context

…And then to get the clean labels, that’s a UX challenge? Correct. Although clean labels, I think maybe it’s worth exploring what that exactly means. I think any given labeling strategy will have some number of assumption to make, about what the user is attempting to do. Those assumptions can be formulated in a loss function, or they can be formulated in terms of heuristics that you might use, to just try to estimate or guesstimate what the user’s trying to do. And what really matters is, how accurate are those assumptions? For example, you might say, “Hey, user, push upwards and follow the speed of this cursor.” And your heuristic might be that they’re trying to do exactly what that cursor is trying to do. Another competing heuristic might be, they’re actually trying to go slightly faster at the beginning of the movement and slightly slower at the end. And those competing heuristics may or may not be accurate reflections of what the user is trying to do. Another version of the task might be, “Hey, user, imagine moving this cursor a fixed offset.” So rather than follow the cursor, just try to move it exactly 200 pixels to the right. So here’s the cursor, here’s the target, okay, cursor disappears, try to move that now invisible cursor, 200 pixels to the right. And the assumption in that case would be that the user can’t actually modulate correctly that position offset. But that position offset assumption might be a weaker assumption, and therefore potentially, you can make it more accurate, than these heuristics that are trying to guesstimate at each millisecond what the user’s trying to do. So you can imagine different tasks that make different assumptions about the nature of the user intention. And those assumptions being correct is what I would think of as a clean label. For that step, what are we supposed to be visualizing? There’s a cursor, and you want to move that cursor to the right, or the left, or up and down, or maybe move them by a certain offset. So that’s one way. Is that the best way to do calibration? So for example, an alternative crazy way that probably is playing a role here, is a game like WEG Grid. Where you’re just getting a very large amount of data, the person playing a game. Where if they’re in a state of flow, maybe you can get clean signal as a side effect?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.