{"id":166044,"date":"2023-01-19T09:48:00","date_gmt":"2023-01-19T08:48:00","guid":{"rendered":"https:\/\/liora.io\/en\/?p=166044"},"modified":"2026-02-06T09:07:40","modified_gmt":"2026-02-06T08:07:40","slug":"recurrent-neural-network-what-is-it","status":"publish","type":"post","link":"https:\/\/liora.io\/en\/recurrent-neural-network-what-is-it","title":{"rendered":"Recurrent Neural Networks: The complete Guide"},"content":{"rendered":"If you are a follower of our blog, you already know what a neural network is (if not, feel free to read this article first) but what does the adjective recurrent bring to this model? In this article, we will see how recurrent neural networks, called RNN, have become a classic model in deep learning<\/a>.<\/strong>

How to set up this neural network in situations ?<\/h3>\nBefore explaining a RNN<\/b>, let’s focus on a bullet. Indeed, it is common in Machine Learning<\/a><\/strong> to want to predict the trajectory of a movable object. As shown in figure 1, from the starting point, the ball can take all sorts of directions<\/b>\"\"\n\nHowever, if we take figure 2<\/b>, it is obvious to say that the ball will continue to go towards the right<\/b>, thanks to the past trajectories, which translate a movement towards the right. \n\nSo far, everything seems logical.\n\nContrary to figure 1, we have more training data, from which we are better able to decide on the movement of the ball.\n\nThus, we just need to give our neural network<\/b> the old ball movements and our study is over. However, how do we choose the number of input neurons? A trajectory can be split in any way we want. Whether it is 10 or 100.\n\n\"\"\n\nThe white ball represents the current position of the ball and the light blue balls represent the old trajectories<\/b> of the white ball, so we guess that the ball is heading to the right. \n\nLet’s ignore this detail for now, and fix the size of our input samples. Let’s look at Figure 3<\/b>, where will the ball go?\n\n\"\"\n\nAdmittedly, the motion of the ball is more complex <\/b>than in figure 1, but we can still tell that the next move of the ball will be upwards<\/b>. Will the model make the same prediction? Unfortunately not, because the neural network does not think like we do! The model, unlike us, does not take into account the link between the inputs. The inputs are not independent<\/b> of each other, so we must preserve this link between them when we train our neural network. \n\nWe need to overcome 2 problems<\/b>: \n