Bayesian Neural Networks

Bayesian Neural Networks uses Proabibilitic Model along with Neural Networks.

• Probabilistic Programming Libraries:
  • PyMC3, Edward, Stan
• Bayes Law
  $Posterior = \frac {likelihood * prior}{Normalizing Constant }$

References:

1. Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning.
2. Loss-Calibrated Approximate Inference in Bayesian Neural Networks.
3. Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding
4. Weight Uncertainty in Neural Networks
5. Study of Bayesian Neural Networks