4 ms·
> For these methods, it however turned out that 6 of them can often be outperformed with comparably simple heuristic methods, e.g., based on nearest-neighbor or
by derefr 7y ago
> For these methods, it however turned out that 6 of them can often be outperformed with comparably simple heuristic methods, e.g., based on nearest-neighbor or graph-based techniques.
Even if these models are outperformed, would they be useful anyway, to serve as insights into the functioning of an agent that can deduce solutions to these tasks "from scratch?"
After all, the human brain is a wondrous AI, but it is often also outperformed by simple heuristics.
- desc 7y agoIt's outperformed at basic arithmetic by fairly simple electronic circuits too. Which is probably your point :P Right tool for the right job. The problem is that when you have a hammer called machine learning, every problem looks like a machine-learning-sized nail.