Interview with Doug Duhaime, contributor to Google’s Dev Library


Posted by the Google Dev Library Crew

Introducing the Dev Library Contributor Spotlights – a weblog sequence highlighting builders which are supporting the thriving improvement ecosystem by contributing their assets and instruments to Google Dev Library.

We met with Doug Duhaime, Full Stack Developer in Yale College’s Digital Humanities Lab, to debate his ardour for Machine Studying, his processes and what impressed him to launch his PixPlot challenge as an Open Supply.

What led you to discover the sector of machine studying?

I used to be an English main in undergrad and in graduate faculty. I’ve a PhD in English literature. My dissertation was exploring copyright historical past and the ways in which adjustments in copyright regulation affected the guide market. How does the establishment of mounted length copyright affect the guide market? To reply this query, I needed to mine an unlimited assortment of knowledge – half one million books, printed earlier than 1800 – to have a look at completely different patterns. That was one of many key tasks that acquired me impressed to additional discover the world of Machine Studying.

In truth, one in all my tasks – the PixPlot library – makes use of laptop imaginative and prescient to investigate picture collections, which was additionally partially utilized in my analysis. A part of my analysis checked out plagiarism detection and the way readily individuals are inclined to repeat pictures as soon as it turns into authorized to repeat them from different texts. Pc imaginative and prescient helps us to reply these questions and determine key patterns.

I’ve seen machine studying and programming as a method to ask new questions in historic contexts. And there is a complete discipline of us – we’re referred to as digital humanists. Yale College, the place I have been for the final 5 years, has a implausible digital humanities program the place researchers are asking questions like this and utilizing enjoyable machine studying platforms like TensorFlow to reply these questions.

Screenshot from the PixPlot library showing Image Fields in the Meserve-Kunhardt Collection with the following identified hotspots: Boxers, Buildings, Buttons, Chairs, Gowns

Are you able to inform us extra concerning the evolution of your PixPlot library challenge?

We began in Yale’s digital humanities lab with a challenge referred to as neural neighbors. And the concept right here was to seek out patterns within the Meserve-Kunhardt Assortment of pictures.

Meserve-Kunhardt is a group of pictures largely from the nineteenth century that Yale just lately acquired. After being acquired by the college, some curators had been getting ready to determine all this actually wealthy metadata to explain these pictures. Nonetheless, they’d a backlog, and so they wanted assist to attempt to make sense of what is on this assortment. And so, Neural Neighbors was our preliminary try to reply this query.

As this challenge went on, we began operating up towards limitations and asking greater questions. For instance, as an alternative of simply trying on the footage, what would it not be like to have a look at your entire assortment ? With the intention to reply this query, we would have liked a extra performant rendering layer.

So we determined to make the most of TensorFlow, which allowed us to extract vector illustration of every picture. We then compressed the dimensionality of these vectors all the way down to 2D. However for PixPlot, we determined to make use of a unique dimensionality discount method referred to as umap. And that introduced us to the primary launch of PixPlot.

The thought right here was to take the entire assortment, shoot it down into 2D, after which allow you to transfer via it and take a look at the photographs within the assortment whereby we count on pictures with comparable content material to be positioned shut by each other.

And so it is simply developed from that early genesis and Neural Neighbors via to the place it’s at the moment.

What impressed you to launch PixPlot as an open supply challenge?

Within the case of PixPlot, I used to be working for Yale College, and we had a purpose to make as a lot of our contributions to the software program world as doable open and publicly accessible with none business phrases.

It was an enormous privilege to spend time with the lab and construct software program that others discovered helpful. I might say much more usually, in my private life, I actually like constructing issues that folks discover helpful and, when doable, contributing again to the open supply world as a result of, I feel, so many people be taught from open supply.

Google Dev Library Quote: We look at other people's examples and get excited by tools and projects others are building. And many of those are non-commercial. They're just open and free to the world. And it's great to give back when we can. Doug Duhaime Dev Library Contributor

Discover out extra content material contributed and authored by Doug Duhaime and uncover extra distinctive instruments and assets on the Google Dev Library web site!


Leave a Reply