Bookkeeping in Machine Learning Experiments

[jaidevd]

• Abstract (2-3 lines)

Solving a machine learning problem is like traversing a minefield, where the safest path can only be determined by blowing up a significantly large number of mines. You can only figure out the right approach after making a bunch of mistakes. Since there is no general rule for determining a 'best model', most things in deep learning can only be solved with trial and error. To a large extent, this 'see what sticks' approach cannot be avoided. However it can be curbed significantly, with a structured approach to running machine learning experiments. This structured approach is what this talk is about.

This talk will introduce a lab journal powered by Python, and optimized for deep learning experiments. It will allow users to log experiments carried out on sklearn estimators and keras models. The journal also behaves like a hyperparameter grid manager, which also alerts the user if the user accidentally re-runs the same experiment on the same data with the same parameters. It will have some meta-learning features which allow for an end-to-end approach to machine learning experiments.

• Pre-requisites for the talk

• Knowledge of basic ML terminology

• Time required for the talk - 20 to 30 minutes

• Link to slides (TBA)

• Will you be doing hands-on demo as well? No

• About yourself

I'm a senior data scientist at Gramener. I build data-driven products and the tooling around them for a living. My research interests are in signal processing and computational harmonic analysis. I'm obsessed with applications of machine learning in personal productivity and recommendation systems. I blog about these here.

• Are you comfortable if the talk is recorded and uploaded to PyData Delhi's YouTube channel ? Yes

[shagunsodhani]

Thanks for submitting the talk @jaidevd . It looks pretty interesting. Best of luck :)