- cross-validate everything
- if an algorithm has tunable parameters, create precision-recall graphs (with cross validation these should give nice graphs with uncertainty regions)
- look at failure modes of each algorithm and compare to others, this will require communication between us
- structure of tex document: introduction, theory of algorithms, data exploration, results of algorithms with comparison, closing notes
- Baye's Classifier
- LDA (Neish)
- QDA (Neish)
- (k-) nearest Neighbour (Jannik)
Logistic Regression(is the same as QDA for binary classification problem)- Histograms
- Density Trees
- SVM (linear and kernel)
- Boosting / Ensemble methods? random forests?
- Multilayer Perceptron
- http://scikit-learn.org/stable/auto_examples/cluster/plot_cluster_comparison.html#sphx-glr-auto-examples-cluster-plot-cluster-comparison-py
- http://scikit-learn.org/stable/auto_examples/neural_networks/plot_mlp_alpha.html#sphx-glr-auto-examples-neural-networks-plot-mlp-alpha-py
- http://scikit-learn.org/stable/auto_examples/svm/plot_rbf_parameters.html#sphx-glr-auto-examples-svm-plot-rbf-parameters-py