Further reading

• The Unreasonable Effectiveness of Recurrent Neural Networks: An excellent article on RNNs (karpathy.github.io/2015/05/21/rnn-effectiveness/)
• Understanding softmax and the negative log-likelihood: Complimentary information on log-likelihood (ljvmiranda921.github.io/notebook/2017/08/13/softmax-and-the-negative-log-likelihood/)
• Finding Structure in Time: An original paper (1990) on RNNs (crl.ucsd.edu/~elman/Papers/fsit.pdf)
• Gradient-Based Learning Applied to Document Recognition: An original paper (1998) on CNNs (yann.lecun.com/exdb/publis/pdf/lecun-98.pdf)
• The Neural Network Zoo: An amazing list of neural network architectures that you can refer to throughout this book (asimovinstitute.org/neural-network-zoo/)