Adrian - - London
1st lesson free
Adrian - - London

Adrian's email address and phone number have been validated.

Adrian

  • Rate ₦107843
  • Response 7h
  • Students

    Number of students accompanied by Adrian since their arrival at Superprof

    3

    Number of students accompanied by Adrian since their arrival at Superprof

Adrian - - London
  • 5 (1 review)

₦107843/h

1st lesson free

Contact

1st lesson free

1st lesson free

Machine learning and data science with explanations and codes for beginners up to computer science or statistics undergraduate and graduate students

Lesson location

    • webcam

About Adrian

I have been teaching in top universities for over 20 years and am a senior professor. I have taught and/or authored papers on these topics. You will receive dedicated teaching, full explanations and help instructions on how to implement the methods you learn. Let me know your level and the lessons can be designed accordingly, either for pure beginners or more advanced learners.

See more

About the lesson

  • All Levels
  • English

All languages in which the lesson is available :

English

Do you want to understand machine learning either as a beginner or an intermediate/advanced student? These lessons can start at the appropriate level, depending on your existing knowledge. Included for free will be codes in Python and instructions on how to run and adapt them. Topics that can be covered include (but are not limited to) - Background in linear algebra, vector calculus, probability, random variables/vectors and random processes - Background in convex optimisation, optimisation for machine learning, stochastic gradient descent (momentum, AdaGrad, Nesterov Accelerated Gradient, RMSProp and Adam) - Linear regression, regularisation (ridge, lasso, Elastic Net; training via proximal gradient methods and coordinate descent); maximum likelihood approaches; basis function methods; multi-output regression; bias-variance decomposition; methods to prevent overfitting; measures of error; cross validation, grid searches and nested cross validation; information criteria and Kullback-Liebler divergence for model selection; nonlinear least squares; Bayesian linear regression; maximum a-posterior estimates; connections to regularisation; evidence approximation and Laplace approximation - First look at non-parametric regression approaches; including local linear and polynomial regression and scatterplot smoothing; k-nearest neighbour regression; decision tree regression and cost complexity pruning; additive and generalised additive models - Linear classification:linear discriminant approaches; probabilistic generative models including naive Bayes classifier; maximum likelihood in classification models; Fisher's discriminant analysis; logistic regression; least-squares classification; error measures; generalised linear models; multi-class problems - Kernel methods and sparse kernel machines for regression: dual representations and kernel ridge regression; the Nadaraya-Watson model; support vector and relevance vector regression; Gaussian process models, including point estimates, mean-field variational inference and Markov Chain Monte Carlo sampling. Also covered are multi-output Gaussian process models, including coregionalisation models and low-rank approximations - An introduction to neural networks (multi-layer perceptrons MLPs) ; deep learning with MLPs, convolutional networks (CNNs); recurrent networks (RNNs); graph networks (GNNs); encoder-decoder models; auto-encoders; training stochastic gradient descent and backpropagation - Nonlinear classification: neural network classifiers, maximal margin classifiers and kernel support vector machines, relevance vector classification; decision tree classifiers; kNN classifiers; Gaussian process classification - Unsupervised learning: dimension reduction, clustering, density estimation: - principal component analysis, multi-dimensional scaling and independent component analysis; Isomap, diffusion maps, kernel PCA and local linear embedding; - clustering using k-means and k-mediod, hierarchical clustering, spectral clustering and Gaussian mixture models with expectation-maximization - density estimation using histograms, kernel density estimation, Gaussian mixture models, the Parzen window method and mixtures of experts - Advanced topics: ensemble methods: bagging (bootstrap aggregating) and boosting, including gradient boosting machines/random forests; - Time series and sequence data using embeddings, supervised machine learning, autoregression via linear/nonlinear state-space models and RNNs

See more

Rates

Rate

  • ₦107843

Pack prices

  • 5h: ₦539217
  • 10h: ₦1078434

free lessons

This first lesson offered with Adrian will allow you to get to know each other and clearly specify your needs for your next lessons.

  • 1hr

webcam

  • ₦107843/h

Similar tutors in London

SuperprofPrivate lessons
See all tutors

Tutor profile verified

Adrian's email address and phone number have been validated.

levels

All Levels

All languages in which the lesson is available

English

Review

All reviews are collected by us and are 100% reliable. They correspond to a real experience of Adrian students

Reviews

The ratings include all certified reviews

as well as the recommendations left on Adrian

Recommendations

Recommendations come from relatives, friends and acquaintances of the tutor

free lessons

This first lesson offered with Adrian will allow you to get to know each other and clearly specify your needs for your next lessons.

Report this ad

Describe the problem encountered on this ad

About
All subjects
Cities
Blogs & Help