Information121 ELU (Exponential Linear Unit) Information The ELU (Exponential Linear Unit) is a type of activation function commonly used in artificial neural networks. It is similar to the ReLU activation function, but with some modifications to handle negative inputs better. The ELU function is defined as f(x) = x for x ≥ 0 and f(x) = α(e^x - 1) for x < 0, where α is a hyperparameter that controls the output value for negative inputs. Th.. Zettelkasten/Terminology Information 2023. 2. 27. MAPE (Mean Absolute Percentage Error) Information MAPE (Mean Absolute Percentage Error) is a measure of the accuracy of a forecasting model, expressed as a percentage of the actual values. MAPE is defined as the average of the absolute percentage errors (APEs) over the forecast horizon, multiplied by 100% to express it as a percentage: MAPE = (1/n) * sum(|(actual - forecast)/actual|) * 100% APE measures the size of the errors in the.. Zettelkasten/Terminology Information 2023. 2. 26. PReLU (Parametric Rectified Linear Unit) Information PReLU (Parametric Rectified Linear Unit) is a variation of the ReLU activation function used in neural networks. It is called "parametric" because it has a learnable parameter that can be adjusted during the training process, unlike the standard ReLU function. The PReLU function is defined as f(x) = alpha * x for x = 0, where alpha is a learnable parameter. Th.. Zettelkasten/Terminology Information 2023. 2. 26. CVRMSE (Coefficient of Variation of the Root Mean Squared Error) Information CVRMSE (Coefficient of Variation of the Root Mean Square Error) is a measure of the variation of the errors in a regression model, normalized by the mean of the target variable. CVRMSE is defined as the ratio of the root mean square error (RMSE) to the mean of the target variable, multiplied by 100% to express it as a percentage: CVRMSE = (RMSE / mean(target)) * 100% RMSE measures th.. Zettelkasten/Terminology Information 2023. 2. 25. LReLU (Leaky Rectified Linear Unit) Information LReLU (Leaky Rectified Linear Unit) is a type of activation function used in deep learning models, particularly in convolutional neural networks (CNNs). It is similar to the ReLU (Rectified Linear Unit) activation function, but it allows for a small, non-zero gradient when the input is negative. The LReLU function is defined as f(x) = max(ax, x), where a is a small constant that is u.. Zettelkasten/Terminology Information 2023. 2. 25. SELU (Scaled Exponential Linear Unit) Information SELU (Scaled Exponential Linear Unit) is an activation function for neural networks that was introduced in 2017 by Klambauer et al. SELU is a self-normalizing activation function, which means that it preserves the mean and variance of the activations across the layers, and thus reduces the vanishing/exploding gradients problem. SELU is defined as a piecewise function that is similar .. Zettelkasten/Terminology Information 2023. 2. 24. Gradient vanishing Information Gradient vanishing refers to a problem that occurs during the training of deep neural networks where the gradients used to update the model's parameters become extremely small as they propagate through the layers of the network. This happens because gradients are calculated using the chain rule of differentiation, and the chain rule involves multiplying many small gradients together,.. Zettelkasten/Terminology Information 2023. 2. 24. Bagging (Bootstrap Aggregating) Information Bagging (Bootstrap Aggregating) is a machine learning technique that combines multiple models trained on different subsets of the training data. Bagging is often used to reduce the variance and improve the stability of the predictions. Bagging samples the training data with replacement to create multiple bootstrap samples, each of which has the same size as the original dataset. Bagg.. Zettelkasten/Terminology Information 2023. 2. 23. Long-term dependency Information In machine learning and artificial neural networks, long-term dependency refers to the challenge of capturing relationships between input and output variables that are separated by a significant time gap. Long-term dependencies can be particularly important in time series forecasting and natural language processing tasks, where the input and output sequences can be very long and comp.. Zettelkasten/Terminology Information 2023. 2. 23. Boosting Information Boosting is a machine learning technique that combines multiple weak learners to create a strong learner. Weak learners are models that perform only slightly better than random guessing, such as decision trees with limited depth or simple linear models. Boosting iteratively trains a sequence of weak learners, where each subsequent model focuses on the samples that were misclassified .. Zettelkasten/Terminology Information 2023. 2. 22. MSE (Mean Squared Error) Information MSE (Mean Squared Error) is a commonly used metric to evaluate the performance of a machine learning model. It measures the average squared difference between the predicted values and the actual values. To calculate the MSE, you take the sum of the squared differences between the predicted and actual values, and then divide by the number of data points. The formula for MSE is: (1/n) .. Zettelkasten/Terminology Information 2023. 2. 22. DNN (Deep Neural Network) Information DNN (Deep Neural Network) is a type of artificial neural network with multiple layers between the input and output layers. DNNs are used for various machine learning tasks, such as image and speech recognition, natural language processing, and autonomous systems. DNNs use backpropagation, a supervised learning algorithm, to adjust the weights of each layer to minimize the difference .. Zettelkasten/Terminology Information 2023. 2. 21. Autoformer Information Autoformer is a variation of the Transformer architecture designed for long-term series forecasting. Autoformer uses an autoregressive model, which means that it uses previous values of the time series as inputs to predict future values. In addition to the standard self-attention mechanism in the Transformer, Autoformer introduces an auto-correlation attention mechanism, which consid.. Zettelkasten/Terminology Information 2023. 2. 21. RF (Random Forest) Information RF (Random Forest) is a machine learning algorithm used for supervised learning problems, such as classification and regression. RF is an ensemble learning method that combines multiple decision trees to improve prediction accuracy. RF builds multiple decision trees using random subsets of the features and data, in order to reduce overfitting and improve generalization. RF uses a com.. Zettelkasten/Terminology Information 2023. 2. 20. DWT (Discrete Wavelet Transform) Information The DWT (Discrete Wavelet Transform) is a signal processing technique used for time-frequency analysis of signals. The DWT decomposes a signal into a set of wavelets, which are small waves with specific properties such as frequency and time localization. The DWT has advantages over other signal processing techniques, such as Fourier analysis, because it can analyze non-stationary sig.. Zettelkasten/Terminology Information 2023. 2. 20. XGBoost (eXtreme Gradient Boosting) Information XGBoost (eXtreme Gradient Boosting) is a machine learning algorithm used for supervised learning problems, such as classification and regression. XGBoost is an ensemble learning method that combines multiple decision trees to improve prediction accuracy. XGBoost uses gradient boosting to iteratively train decision trees in a way that minimizes the overall prediction error. XGBoost us.. Zettelkasten/Terminology Information 2023. 2. 19. DFT (Discrete Fourier Transform) Information DFT (Discrete Fourier Transform) is a mathematical technique used for frequency analysis of a finite set of discrete data. DFT is a transformation of a discrete signal from the time domain to the frequency domain. It decomposes a signal into a sum of sinusoids of different frequencies. DFT is widely used in digital signal processing, image processing, and data compression. DFT is com.. Zettelkasten/Terminology Information 2023. 2. 19. CCHP (Combined, cooling, heating and power) Information CCHP (Combined cooling, heating, and power) is an integrated energy system that produces electricity, heating, and cooling simultaneously from a single fuel source. CCHP systems can be powered by various fuel sources, including natural gas, biogas, and biomass. CCHP systems are highly efficient, with overall energy efficiency rates exceeding 80%, compared to around 50% for traditiona.. Zettelkasten/Terminology Information 2023. 2. 18. LR (Linear Regression) Information LR (Linear Regression) is a statistical technique used to model the relationship between one or more independent variables and a dependent variable. LR assumes a linear relationship between the independent and dependent variables, meaning that the change in the dependent variable is proportional to the change in the independent variable. LR aims to find the line of best fit that mini.. Zettelkasten/Terminology Information 2023. 2. 17. Load forecasting Information Load forecasting is the process of estimating future electricity demand based on historical and current data, as well as other relevant factors such as weather patterns and economic trends. Load forecasting is essential for utilities and energy providers to plan for the future and ensure a reliable and efficient electricity supply. Load forecasting can be short-term (up to 24 hours),.. Zettelkasten/Terminology Information 2023. 2. 16. 이전 1 ··· 3 4 5 6 7 다음