Simulated Annealing is a stochastic global search optimization algorithm. This means that it makes use of randomness as part of the search process. This makes the algorithm appropriate for nonlinear objective functions where other local search algorithms do not operate well. Like the stochastic hill climbing local search algorithm, it modifies a single solution and […]
The No Free Lunch Theorem is often thrown around in the field of optimization and machine learning, often with little understanding of what it means or implies. The theorem states that all optimization algorithms perform equally well when their performance is averaged across all possible problems. It implies that there is no single best optimization […]
Stochastic optimization refers to the use of randomness in the objective function or in the optimization algorithm. Challenging optimization algorithms, such as high-dimensional nonlinear objective problems, may contain multiple local optima in which deterministic optimization algorithms may get stuck. Stochastic optimization algorithms provide an alternative approach that permits less optimal local decisions to be made […]
It can be challenging to develop a neural network predictive model for a new dataset. One approach is to first inspect the dataset and develop ideas for what models might work, then explore the learning dynamics of simple models on the dataset, then finally develop and tune a model for the dataset with a robust […]
Regression models are fit on training data using linear regression and local search optimization algorithms. Models like linear regression and logistic regression are trained by least squares optimization, and this is the most efficient approach to finding coefficients that minimize error for these models. Nevertheless, it is possible to use alternate optimization algorithms to fit […]
Optimization involves finding the inputs to an objective function that result in the minimum or maximum output of the function. The open-source Python library for scientific computing called SciPy provides a suite of optimization algorithms. Many of the algorithms are used as a building block in other algorithms, most notably machine learning algorithms in the […]
Gradient descent is an optimization algorithm that follows the negative gradient of an objective function in order to locate the minimum of the function. A problem with gradient descent is that it can bounce around the search space on optimization problems that have large amounts of curvature or noisy gradients, and it can get stuck […]
Weight initialization is an important design choice when developing deep learning neural network models. Historically, weight initialization involved using small random numbers, although over the last decade, more specific heuristics have been developed that use information, such as the type of activation function that is being used and the number of inputs to the node. […]
There is a lot of confusion for beginners around what algorithm is used to train deep learning neural network models. It is common to hear neural networks learn using the “back-propagation of error” algorithm or “stochastic gradient descent.” Sometimes, either of these algorithms is used as a shorthand for how a neural net is fit […]
Optimization refers to finding the set of inputs to an objective function that results in the maximum or minimum output from the objective function. It is common to describe optimization problems in terms of local vs. global optimization. Similarly, it is also common to describe optimization algorithms or search algorithms in terms of local vs. […]
