Category: Deep Learning
What is ReLU and Sigmoid activation function?
The activation function is a nonlinear function that takes in the weighted sum and produces the output. They are used to provide a more simplified model of neuron behavior which can be used as an input to deep neural networks. There are many different activation functions that can be used, including sigmoid, hyperbolic tangent, logistic,…
Read MoreWhat is Vanishing and exploding gradient descent?
Vanishing and exploding gradient descent is a type of optimization algorithm used in deep learning. Vanishing Gradient Vanishing Gradient occurs when the gradient is smaller than expected. It causes the earlier layers to start degrading before the later ones do, causing a decrease in the overall learning rate of that subset of layers. The weights…
Read MoreWhat is forward and backward propagation in Deep Learning?
Forward propagation is a process in which the network’s weights are updated according to the input, output and gradient of the neural network. In order to update the weights, we need to find the input and output values. The input value is found by taking the difference between the current hidden-state value and that of…
Read MoreDifference between perceptron and neuron?
Perceptrons are a type of artificial neural network that can be used for classification and regression. They are supervised learning algorithms, meaning they need labeled input data in order to learn. how to map inputs to outputs. What independent variables do perceptrons need? Perceptrons require at least one input and one output. What are the…
Read MoreWhat is perceptron?
The perceptron is a type of artificial neural network (ANN) that is designed to recognize patterns in data. It can be used to identify objects, classify images, and detect changes in the environment. The perceptron was invented by Frank Rosenblatt in 1957 while he was working at Cornell Aeronautical Laboratory as part of a research…
Read MoreWhat is L1 and L2 regularization in Deep Learning?
In deep learning, L1 and L2 regularization are regularization techniques used to penalize the model’s weights during the training process. This penalty discourages the model from assigning excessive importance to certain features, thereby reducing the risk of overfitting. L1 Regularization L1 regularization, also known as Lasso regularization, adds a penalty proportional to the absolute value…
Read MoreWhat is the difference between LSTM and GRU?
LSTM or Long Short Term Memory is a kind of Recurrent Neural Network that is capable of learning long-term patterns. It was developed by Schmidhuber and Hochreiter in 1997. It connects sequences of memory in a way that makes it difficult to remember each of the items for an extended period of time. A Globally…
Read MoreDifference between Artificial Intelligence, Machine Learning and Deep Learning?
Artificial intelligence (AI) is a term that encompasses computer systems designed to imitate human intelligence. It is an exciting field that has attracted considerable attention in many industries, including finance, hospitality, education and entertainment. Artificial intelligence is planned to simulate human behavior and thought processes, making it one of the most important trends of this…
Read MoreWhat is semi-supervised Learning?
Semi-supervised learning is a technique in between supervised and unsupervised learning. Arguably, it should not be a category of machine learning but only a generalization of supervised learning, but it’s useful to introduce the concept separately. Its aim is to reduce the cost of gathering labelled data by extending a few labels to similar unlabeled…
Read MoreWhat is LSTM and How does it work?
One of the most common problems in RNNs is called gradient vanishing. LSTM architectures help you with this. A very common type of RNN is LSTM. This type of network is much better at capturing long-term dependencies than simple RNNs. The only unusual thing about LSTMs is the way that they compute the hidden state.…
Read More