{"id":2260,"date":"2020-10-13T16:29:58","date_gmt":"2020-10-14T00:29:58","guid":{"rendered":"https:\/\/live-cometml.pantheonsite.io\/blog\/how-to-start-the-machine-learning-research-process\/"},"modified":"2020-10-13T16:29:58","modified_gmt":"2020-10-14T00:29:58","slug":"how-to-start-the-machine-learning-research-process","status":"publish","type":"post","link":"https:\/\/www.comet.com\/site\/blog\/how-to-start-the-machine-learning-research-process\/","title":{"rendered":"Industry Q&A: Starting the ML Process"},"content":{"rendered":"\n

Comet recently hosted the online panel, \u201cHow do top AI researchers from Google, Stanford and Hugging Face approach new ML problems?\u201d<\/a> This post is our first in a series where we recap the questions, answers, and approaches that top AI teams in the world are taking to critical machine learning challenges. <\/em><\/p>\n\n\n\n

We would like to thank Ambarish Jash<\/a>, Google<\/a><\/em>; Piero Molino<\/a>, Stanford<\/a><\/em> + Ludwig<\/a><\/em>; and Victor Sanh<\/a>, Hugging Face<\/a><\/em>; for their participation. <\/em><\/p>\n\n\n\n

\n
https:\/\/vimeo.com\/466584783<\/div>\n<\/figure>\n\n\n\n

One of the hardest parts of machine learning is simply getting started. Do you have the necessary data? Do you have the systems in place to manage your model? If you need to take it into production, do you have a good understanding of what the production environment looks like? All of these are considerations that need to be made in order to ensure your work is successful — and if the problem you\u2019re trying to solve is worth pursuing.<\/p>\n\n\n\n

Gideon Mendels, Comet<\/strong>
There are a number of challenges to even approaching a machine learning challenge. There are so many moving parts. For those of us in the industry, it\u2019s very different from what you might see in something like a Kaggle competition, where you have a clean dataset and the metrics are figured out for you.<\/p>\n\n\n\n

So how do you start the research process? What do you do when you have a new problem?<\/p>\n\n\n\n

Ambarish Jash, Google AI<\/strong>
Two parts, one is the research problem and coming up with the problem definition. The other is the challenge of putting it in production. Production puts a significant amount of constraint on the research you can do. One approach is to:<\/p>\n\n\n\n