The weird thing about Gary Marcus's constant critique of deep learning is that the most prominent researchers in the field actually agree with him (and are doin

The weird thing about Gary Marcus's constant critique of deep learning is that the most prominent researchers in the field actually agree with him (and are doing something about it), yet he casts himself as a maverick, and that plays well in the media. Here's an example of real researchers making real progress combining DL with other approaches to AI: https://openreview.net/pdf?id=rJgMlhRctm https://openreview.net/pdf?id=rJgMlhRctm