Four short links: 17 April 2018
Dubsteganography, Parsing History, Hackin' the Jack In, and Model Bias
- Hide Data in Dubstep Drops — the blog post shows how to use it. Skrillex meets steganography!
- Parsing Timeline — wonderfully detailed, yet it reads almost chatty. Interesting and informative.
- Securing Wireless Neurostimulators — a hack and discussion of the risk of insecure implantable medical devices that interface with the brain. (via Paper a Day)
- Text Embedding Models Contain Bias (Google) — great to see this making its way to research outputs, instead of being the province of damage control and bad PR. The Developers section of the Semantic Experiences microsite talks about “unwanted associations”: In Semantris, the list of words we’re showing are hand curated and reviewed. To the extent possible, we’ve excluded topics and entities that we think particularly invite unwanted associations, or can easily complement them as inputs. In Talk to Books, while we can’t manually vet each sentence of 100,000 volumes, we use a popularity measure which increases the proportion of volumes that are published by professional publishing houses. There are additional measures that could be taken. For example, a toxicity classifier or sensitive topics classifier could determine when the input or the output is something that may be objectionable or party to an unwanted association. We recommend taking bias-impact mitigation steps when crafting end-user applications built with these models.
