4 ms·
It’ll be very hard because it can be impossible to trace a model back to it’s constituent data, especially if it’s ephemeral
by radicaldreamer 8y ago
It’ll be very hard because it can be impossible to trace a model back to it’s constituent data, especially if it’s ephemeral
- yati 8y agoOne way is to make sure the provenance is known from source tuples to all the way to the model version, and (eventually) retrain the model when any records upstream of it are deleted.
- groestl 8y agoAFAIK Palatir made a business model out of this (i.e. tracking provenance for datasets and models, along with permissions). (Just hearsay, not affiliated with Palantir in any way)
- blitmap 8y agoWell see that's the other thing: "Someday" they'll have to make these connections in case they are legally obligated to erase a data source and it's derivatives.
- groestl 8y agoEphemeral is not an option then, isn't it? In face of regulation, some things can't/shouldn't be optimized. It's already the case for more established industries like pharma.