Neural networks get top results in Computer Vision tasks (see MNIST, ILSVRC, Kaggle Galaxy Challenge). They seem to outperform every other approach in Computer Vision. But there are also other tasks:
- Kaggle Molecular Activity Challenge
- Regression: Kaggle Rain prediction, also the 2nd place
- Grasp and Lift 2nd also third place - Identify hand motions from EEG recordings
I'm not too sure about ASR (automatic speech recognition) and machine translation, but I think I've also heard that (recurrent) neural networks (start to) outperform other approaches.
I am currently learning about Bayesian Networks and I wonder in which cases those models are usually applied. So my question is:
Is there any challenge / (Kaggle) competition, where the state of the art are Bayesian Networks or at least very similar models?
(Side note: I've also seen decision trees, 2, 3, 4, 5, 6, 7 win in several recent Kaggle challenges)