4 ms·
I think Ben Evans is a lot more optimistic than I am about about how feasible to train good models from data. By the time you have considered all the caveats, t
by 2sk21 8y ago
I think Ben Evans is a lot more optimistic than I am about about how feasible to train good models from data. By the time you have considered all the caveats, the number of candidates for applying ML greatly decreases.
- arcanus 8y agoI don't disagree that the number of applications of ML/AI is not as universal as the hype would lead one to believe at present. However, don't underestimate that AI/ML could be a secular growth over twenty+ years. From that perspective, marginal (but constant) growth could result from widespread adoption of these methods (with tweaks) to new markets. A VC should necessarily take a relatively long perspective,so your two viewpoints are not necessarily irreconcilable.