Machine Learning - An Algorithmic Perspective
By Stephen R. Marsland. 2009
1,135 citations recorded by the source corpus.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
Citations in the corpus
- Towards Better Analysis of Deep Convolutional Neural Networks (2016, methodology): “In CNNVis, we employ two widely used clustering methods, K-Means [36] (parametric clustering) and MeanShift [8] (nonparametric clustering).”
- Lyapunov-Guided Deep Reinforcement Learning for Stable Online Computation Offloading in Mobile-Edge Computing Networks (2020, methodology): “We then update the parameter of the DNN by minimizing its average cross-entropy loss function LS ( θ ) over the data samples using the Adam algorithm [30]”
- Efficient training of physics‐informed neural networks via importance sampling (2021, methodology): “Second, we show how a piece-wise constant approximation to the loss function, using nearest neighbor search [35] or Voronoi tessellation [36], can be used to approximate the proposal distribution to further improve the convergence behavior of PINNs training.”
- Startups’ Roads to Failure (2018, methodology): “More in depth, we considered, as clustering methods, k-means, Kohonem network and two-step cluster analysis [59].”
- Emerging artificial intelligence methods in structural engineering (2018, background): “ML based models can be predictive or descriptive to achieve knowledge from the data [37,38].”
- Playing tag with ANN: boosted top identification with pattern recognition (2015, methodology): “ to choose weights that minimize this function. We use the backpropagationalgorithm [43], combinedwith gradient-descentminimization. In itssimplest version, the algorithm can be summarized as follows [44]: 1.Initialize the weights of each link to small random values. 3In Eq. (3.1) and below, repeate”
- Seizure detection, seizure prediction, and closed-loop warning systems in epilepsy. (2014, background): “This second step, called classification, might be as simple as thresholding a value or might require models derived from modern machine learning algorithms [51,52].”
- Geological mapping using remote sensing data: A comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information (2014, methodology): “MLA architecture and the statistical distributions of observed data guides the training of classification models, which is usually carried out by minimizing a loss (error) function (Kuncheva, 2004; Marsland, 2009).”
- Survey of Machine Learning Algorithms for Disease Diagnostic (2017, methodology): “Unsupervised learning contains clustering [1].”
- GossipGraD: Scalable Deep Learning using Gossip Communication based Asynchronous Gradient Descent (2018, unknown): “„‘”a„‘”+b„‘”and a „‘”= σ„z ”, where σ is the activation function, and we set n‘ represent the number of layers. The details of the algorithm are presented in Algorithm 1. Algorithm 1 Back Propagation [45] 1: input: Data X 2Rnp and labels Y 2Rnl 2: Compute all z„‘”and a„‘”. 3: δ„n ‘”= „y a„n ‘””σ„z„n”
- An Overview on Application of Machine Learning Techniques in Optical Networks (2018, background): “MACHINE learning (ML) is a branch of Artificial Intelligence that pushes forward the idea that, by giving access to the right data, machines can learn by themselves how to solve a specific problem [1].”
- Deep Reinforcement Learning for Online Computation Offloading in Wireless Powered Mobile-Edge Computing Networks (2018, background): “The well-known universal approximation theorem claims that one hidden layer with enough hidden neurons suffices to approximate any continuous mapping f if a proper activation function is applied at the neurons, e.g., sigmoid, ReLu, and tanh functions [29].”
- Data Center Energy Consumption Modeling: A Survey (2016, background): “Decision tress can be easily turned into a set of if-then rules which are suitable for use in a rule induction system [365].”
- A Survey on Machine Learning Techniques for Cyber Security in the Last Decade (2020, background): “Instead, the algorithm has to figure out and try several possibilities until it learns the correct answer [195].”