Recurrent Neural Network (RNN)

YouTube Search ...Google search

• Neural Network Zoo | Fjodor Van Veen
• Sequence to Sequence (Seq2Seq)
• Attention Mechanism/Transformer Model: Attention Is All You Need
• Gradient Descent Optimization & Challenges
• A Beginner's Guide to LSTMs and Recurrent Neural Networks | Chris Nicholson - A.I. Wiki pathmind
• Handwriting generation demo | Alex Graves
• Recurrent Neural Network (RNN) Variants:
  • Long Short-Term Memory (LSTM)
  • Gated Recurrent Unit (GRU)
  • Bidirectional Long Short-Term Memory (BI-LSTM)
  • Bidirectional Long Short-Term Memory (BI-LSTM) with Attention Mechanism
  • Average-Stochastic Gradient Descent (SGD) Weight-Dropped LSTM (AWD-LSTM)
  • Hopfield Network (HN)
• Natural Language Processing (NLP)
• AI-Powered Search
• Assistants
• Memory Networks
• Sentiment Analysis | Stanford’s Sentiment Analysis Demo using Recursive Neural Networks
• Generative Modeling
• Animated RNN, LSTM and GRU | Raimi Karim - Towards Data Science
• How Wikimedia is using machine learning to spot missing citations | Seth Colander - VentureBeat
• The Unreasonable Effectiveness of Recurrent Neural Networks | Andrej Karpathy - Towards Data Science
• Visualizing A Neural Machine Translation Model (Mechanics of Seq2seq Models With Attention) | Jay Alammar

Recurrent nets are a type of artificial neural network designed to recognize patterns in sequences of data, such as text, genomes, handwriting, the spoken word, or numerical times series data emanating from sensors, stock markets and government agencies. They are arguably the most powerful and useful type of neural network, applicable even to images, which can be decomposed into a series of patches and treated as a sequence. Since recurrent networks possess a certain type of memory, and memory is also part of the human condition, we’ll make repeated analogies to memory in the brain. Recurrent neural networks (RNN) are FFNNs with a time twist: they are not stateless; they have connections between passes, connections through time. Neurons are fed information not just from the previous layer but also from themselves from the previous pass. This means that the order in which you feed the input and train the network matters: feeding it “milk” and then “cookies” may yield different results compared to feeding it “cookies” and then “milk”. One big problem with RNNs is the vanishing (or exploding) gradient problem where, depending on the activation functions used, information rapidly gets lost over time, just like very deep FFNNs lose information in depth. Intuitively this wouldn’t be much of a problem because these are just weights and not neuron states, but the weights through time is actually where the information from the past is stored; if the weight reaches a value of 0 or 1 000 000, the previous state won’t be very informative. RNNs can in principle be used in many fields as most forms of data that don’t actually have a timeline (i.e. unlike sound or video) can be represented as a sequence. A picture or a string of text can be fed one pixel or character at a time, so the time dependent weights are used for what came before in the sequence, not actually from what happened x seconds before. In general, recurrent networks are a good choice for advancing or completing information, such as autocompletion. Elman, Jeffrey L. “Finding structure in time.” Cognitive science 14.2 (1990): 179-211.

Bidirectional Recurrent Neural Network (BiRNN) look exactly the same as its unidirectional counterpart. The difference is that the network is not just connected to the past, but also to the future. Schuster, Mike, and Kuldip K. Paliwal. “Bidirectional recurrent neural networks.” IEEE Transactions on Signal Processing 45.11 (1997): 2673-2681.