3 ms·
Historically the XOR function has been the simple example that many ML algorithms can't handle. Just imagine a higher dimensional XOR with outliers, and you ha
by CuriouslyC 6y ago
Historically the XOR function has been the simple example that many ML algorithms can't handle. Just imagine a higher dimensional XOR with outliers, and you have a pretty good use case for DL with limited predictors.
- dr_zoidberg 6y agoHistorically, this was solved in the 80's with the multi-layer perceptron, but it seems it still gets repeated.