Cynthia Rudin wins the 2021 AAAI Squirrel AI Award

Post by qPM9l3XJrF » Tue, Oct 12, 2021, 7:06 PM UTC

randcraw wrote: post #38

qPM9l3XJrF wrote: post #7

Earlier quoted context omitted.

Cynthia has a great paper where she distinguishes between "interpretable" and "explainable" https://arxiv.org/pdf/1811.10154.pdf

"Explainable" just means you didn't build your app/service using any technology that isn't interpretable. Her argument is that the strategy of reverse engineering a method to convert it from inexplicable to explicable is inherently less effective than maintaining explicability at all times in the app's genesis -- from the design phase through implementation. But Rudin's Premise is philosophical more than practical. I…

Post by pjmorris » Tue, Oct 12, 2021, 8:18 PM UTC

randcraw wrote: post #38

qPM9l3XJrF wrote: post #7

Earlier quoted context omitted.

Cynthia has a great paper where she distinguishes between "interpretable" and "explainable" https://arxiv.org/pdf/1811.10154.pdf

"Explainable" just means you didn't build your app/service using any technology that isn't interpretable. Her argument is that the strategy of reverse engineering a method to convert it from inexplicable to explicable is inherently less effective than maintaining explicability at all times in the app's genesis -- from the design phase through implementation. But Rudin's Premise is philosophical more than practical. I…

  Here is the Rashomon set argument: Consider that the data permit a large set of reasonably accurate predictive models to exist. Because this set of accurate models is large, it often contains at least one model that is interpretable. This model is thus both interpretable and accurate. 

  Unpacking this argument slightly, for a given data set, we define the Rashomon set as the set of reasonably accurate predictive models (say within a given accuracy from the best model accuracy of boosted decision trees). Because the data are finite, the data could admit many close-to-optimal models that predict differently from each other: a large Rashomon set. I suspect this happens often in practice because sometimes many different machine learning algorithms perform similarly on the same dataset, despite having different functional forms (e.g., random forests, neural networks, support vector machines). As long as the Rashomon set contains a large enough set of models with diverse predictions, it probably contains functions that can be approximated well by simpler functions, and so the Rashomon set can also contain these simpler functions. Said another way, uncertainty arising from the data leads to a Rashomon set, a larger Rashomon set probably contains interpretable models, thus interpretable accurate models often exist.


  If this theory holds, we should expect to see interpretable models exist across domains. These interpretable models may be hard to find through optimization, but at least there is a reason we might expect that such models

Post by randcraw » Tue, Oct 12, 2021, 8:53 PM UTC

qPM9l3XJrF wrote: post #41

randcraw wrote: post #38

Earlier quoted context omitted.

"Explainable" just means you didn't build your app/service using any technology that isn't interpretable. Her argument is that the strategy of reverse engineering a method to convert it from inexplicable to explicable is inherently less effective than maintaining explicability at all times in the app's genesis -- from the design phase through implementation. But Rudin's Premise is philosophical more than practical. I…

>If the problem at hand is better solved using a black box (in terms of accuracy, precision, robustness, etc) It's been a while since I read her work, but IIRC one of the positions she argues for, which I find plausible, is that interpretable models can be performance competitive. For example, it could be that the only reason black box methods outperform is because they've been more heavily researched, and if we were…

Post by typest » Tue, Oct 12, 2021, 9:07 PM UTC

gwf wrote: post #35

I was also one of Cynthia's Ph.D. advisors when she was a graduate student at Princeton, some twenty years ago. It was obvious to me then that she would go on to do great things, so it's delightful to read this news this morning. My fondest memory of Cynthia, however, has nothing to do with science, and everything to do with just being a kind person. We were at the NEC Research Institute's company picnic where they h…

Post by paranoidroid » Tue, Oct 12, 2021, 9:32 PM UTC

paranoidroid wrote: post #48

dr_dshiv wrote: post #46

What’s with “Squirrel AI”? Did that seem slightly out of place?

Squirrel AI is a Chinese online education company. It is the first large scale AI-powered adaptive education provider in China [...] https://en.wikipedia.org/wiki/Squirrel_AI Thousand Talents Obfuscation Initiative?