Automated Machine Learning on Big Data using Stochastic Algorithm Tuning

By T. Nickson, Michael A. Osborne, Steven Reece, Stephen J. Roberts. 2014

25 citations recorded by the source corpus.

We introduce a means of automating machine learning (ML) for big data tasks, by performing scalable stochastic Bayesian optimisation of ML algorithm parameters and hyper-parameters. More often than not, the critical tuning of ML algorithm parameters has relied on domain expertise from experts, along with laborious hand-tuning, brute search or lengthy sampling runs. Against this background, Bayesian optimisation is finding increasing use in automating parameter tuning, making ML algorithms accessible even to non-experts. However, the state of the art in Bayesian optimisation is incapable of scaling to the large number of evaluations of algorithm performance required to fit realistic models to complex, big data. We here describe a stochastic, sparse, Bayesian optimisation strategy to solve this problem, using many thousands of noisy evaluations of algorithm performance on subsets of data in order to effectively train algorithms for big data. We provide a comprehensive benchmarking of possible sparsification strategies for Bayesian optimisation, concluding that a Nystrom approximation offers the best scaling and performance for real tasks. Our proposed algorithm demonstrates substantial improvement over the state of the art in tuning the parameters of a Gaussian Process time series prediction task on real, big data.

Subfields: Artificial Intelligence, Management Science and Operations Research, Optimization & Training, Neural Architecture Search & AutoML

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

In Optimization & Training

In Neural Architecture Search & AutoML

Citations in the corpus

Benchmark and Survey of Automated Machine Learning Frameworks (2019, neutral): “A straight-forward approach to mimic expert behavior is choosing multiple random subsets of the training data for performance evaluation (Nickson et al., 2014).”
Survey on Automated Machine Learning (2019, neutral): “A straight-forward approach to mimic expert behavior is choosing multiple random subsets of the training data for performance evaluation (Nickson et al., 2014).”
Fast Bayesian Optimization of Machine Learning Hyperparameters on Large Datasets (2016, methodology): “In §2 we review Bayesian optimization, in particular the Entropy Search algorithm and the related method of Multi-Task Bayesian optimization.”
Mondrian Forests for Large-Scale Regression when Uncertainty Matters (2015, background): “[20] discuss a scenario where this heuristic produces”
Fast Bayesian hyperparameter optimization on large datasets (2017, unknown): “Other approaches for hyperparameter optimization on large datasets include work by Nickson et al. (2014), who estimated a conﬁguration’s performance on a large dataset by evaluating several training runs on small, random subsets of ﬁxed, manually-chosen sizes.”
Towards efﬁcient Bayesian Optimization for Big Data (2015, background): “Recently, Nickson et al. [5] considered this problem and addressed it by estimating a conﬁguration’s performance by evaluating it based on several training runs on small random subsets of ﬁxed, manually chosen size.”
Taking Human out of Learning Applications: A Survey on Automated Machine Learning (2018, methodology): “• Bayesian optimization [85], [86]: Bayesian optimization builds a probabilistic surrogate function on search space by Gaussian process or other model (e.”