Model selection

Calibration of probabilistic forecasts Proper scoring rules, skill scores etc 2015-06-16 – 2023-11-15

Bitter lessons in compute and cleverness Operationalizing the scaling hypothesis 2021-01-14 – 2023-08-02

Model averaging On keeping many incorrect hypotheses and using them all as one goodish one 2017-06-20 – 2023-07-24

Hypothesis tests, statistical 2014-08-23 – 2023-07-18

Bayesian model calibration 2017-04-11 – 2023-06-01

Canonical correlation 2014-08-23 – 2023-05-17

Hyperparameter optimization Replacing a hyperparameter problem with a hyperhyperparameter problem which feels like progress 2020-09-25 – 2023-05-12

Multi-objective optimisation 2021-07-14 – 2023-05-04

Data summarization On maps drawn at smaller than 1:1 scale 2019-01-14 – 2023-04-27

Leakage in predictive models 2016-09-05 – 2023-04-12

Scaling laws for very large neural nets Compute/size/data tradeoffs 2021-01-14 – 2023-02-16

Bayesian model selection by model evidence maximisation Type II maximum likelihood, marginal maximum likelihood, Bayes Occam’s razor 2017-08-20 – 2022-12-22

Bayesian sparsity 2019-01-08 – 2022-10-25

Randomised linear algebra 2016-08-16 – 2022-10-22

Neural tangent kernel 2020-12-09 – 2022-10-14

Forecasting Vegan haruspicy 2015-06-16 – 2022-10-08

Penalised/regularised regression 2016-06-23 – 2022-09-19

Overparameterization in large models Improper learning, benign overfitting, double descent 2018-04-04 – 2022-05-27

Forecasting with model averaging Mixture of experts, ensembles and time series 2022-05-04

Gradient descent at scale Practical implementation of large optimisations 2021-07-14 – 2021-09-28

Regularising neural networks Generalisation for street fighters 2017-02-12 – 2021-09-24

Fractals and self-similarity 2011-11-13 – 2021-09-22

Meta learning Few-shot learning, learning fast weights, learning to learn 2021-09-16

Sequential experiments Especially multiple sequential experiments 2021-08-04

Random-forest-like methods A selection of randomly stopped clocks is never far from wrong. 2015-09-23 – 2021-06-17

Stein’s method His eyes are like angels but his heart is cold / No need to ask / He’s a Stein operator 2021-03-12 – 2021-06-01

Cross validation 2016-09-05 – 2021-05-13

Infinite width limits of neural networks 2020-12-09 – 2021-05-11

Compressing neural nets pruning, compacting and otherwise fitting a good estimate into fewer parameters 2016-10-14 – 2021-05-07

Differentiable model selection Differentiable hyperparameter search, and architecture search, and optimisation optimisation by optimisation and so on 2020-09-25 – 2021-04-13

Generically approximating probability distributions 2021-03-12 – 2021-03-22

Matrix measure concentration inequalities and bounds 2014-11-25 – 2021-03-08

Measure concentration inequalities The fancy name for probability inequalities 2014-11-25 – 2021-03-04

Weighted data in statistics 2020-11-04 – 2020-11-06

Adaptive design of experiments Minesweeper++ 2017-04-11 – 2020-10-13

Sparse model selection 2016-09-05 – 2020-10-02

AutoML 2017-07-17 – 2020-10-02

Independence, conditional, statistical 2016-04-21 – 2020-09-13

Minimum description length 2020-08-06

Model complexity penalties Information criteria, degrees of freedom etc 2015-04-22 – 2020-06-22

Long memory time series 2011-11-13 – 2020-05-28

Kernel approximation 2016-07-27 – 2020-03-06

Effective sample size 2016-11-21 – 2020-03-03

Convergence of random variables 2019-12-03

Phase retrieval I’ve got the power. / Like the crack of the whip/ I snap attack/ Front to back 2017-01-16 – 2019-11-07

Sparse regression 2016-06-23 – 2019-10-24

Statistical learning theory for time series 2016-11-03 – 2019-10-01

Bayesian model selection 2017-08-20 – 2019-07-22

Wacky regression 2015-09-23 – 2019-05-02

Nearly sufficient statistics How about “Sufficient sufficiency”? — is that taken? 2018-03-13 – 2019-01-14

Multiple testing 2015-04-22 – 2018-11-05

Post-selection inference Adaptive data analysis without cheating 2017-08-20

Model/hyperparameter selection 2016-04-15 – 2017-08-20

Compressed sensing and sampling A fancy ways of counting zero 2014-08-18 – 2017-06-14

Stability (in learning) 2016-05-25 – 2016-10-05

Statistical learning theory Eventually including structural risk minimisation, risk bounds, hopefully-uniform convergence rates, VC-dimension, generalisation-and-stability framings etc 2016-07-06 – 2016-08-16