When DNNs go wrong – adversarial examples and what we can learn from them

Yesterday we looked at a series of papers on DNN understanding, generalisation, and transfer learning. One additional way of understanding what's going on inside a network is to understand what can break it. Adversarial examples are deliberately constructed inputs which cause a network to produce the wrong outputs (e.g., misclassify an input image). We'll start … Continue reading When DNNs go wrong – adversarial examples and what we can learn from them

Understanding, generalisation, and transfer learning in deep neural networks

This is the first in a series of posts looking at the 'top 100 awesome deep learning papers.' Deviating from the normal one-paper-per-day format, I'll take the papers mostly in their groupings as found in the list (with some subdivision, plus a few extras thrown in) - thus we'll be looking at multiple papers each … Continue reading Understanding, generalisation, and transfer learning in deep neural networks

On decentralizing prediction markets and order books

On decentralizing prediction markets and order books Clark et al., 13th Annual Workshop on the Economics of Information Security, 2014 This is the last of five papers in the ACM Queue Research for Practice series on 'Cryptocurrencies, Blockchains, and Smart Contracts .' It serves as a good example of repurposing block chains as a foundation … Continue reading On decentralizing prediction markets and order books

A first look at the usabilty of Bitcoin key management

A first look at the usability of Bitcoin key management Eskandari et al., USEC 2015 This is the third of five papers from the ACM Queue Research for Practice selections on 'Cryptocurrencies, Blockchains, and Smart Contracts.' And thankfully it's much easier to read and understand than yesterdays! The authors point out that a cryptocurrency intended … Continue reading A first look at the usabilty of Bitcoin key management

Zerocash: Decentralized anonymous payments from Bitcoin

Zerocash: Decentralized anonymous payments from Bitcoin Ben-Sasson et al., 2014 Yesterday we saw that de-anonymising techniques can learn a lot about the true identities of participants in Bitcoin transactions. Ben-Sasson et al. point out that given this, Bitcoin could be considered significantly less private than traditional schemes: While users may employ many identities (or pseudonyms) … Continue reading Zerocash: Decentralized anonymous payments from Bitcoin