home | tags | mulvany.net


Ai

Distill is dedicated to making machine learning clear and dynamic

Distill is an experiment in bringing interactive documents and scholarly documents together. I’m often asked what the future of publishing might look like, and were we to embrace what the web offers it might look like distill. Two things though, make it look like a nice product. Right now paper flow into this journal is very low, and secondly they have advertised a large prize to attract work in this format. ... (more)

Narrative Science | Natural Language Generation Technology

I think what these people are doing is creating human readable summaries of large corpora of documents or signals based on AI. At some point this kind of approach is going to mature to the point of being able to replace many traditional editorial functions, but its going to happen in finance first. https://narrativescience.com/ ... (more)

Model Zoo - Pretrained deep learning models for transfer learning, educational purposes, and more

This looks like an interesting corpus of machine learning models. One thing I’d not really thought about before is that we will now need to start thinking about how to publish the actual models that are used in scholarly research. Some of the models on this site provide ways to reference the work (e.g. https://modelzoo.co/model/image-to-image-translation-with-conditional-adversarial-networks). https://modelzoo.co/ ... (more)

Some initial thoughts on the Google Duplex demo

The Verge also covered this well: www.theverge.com/2018/5/9/17334658/google-ai-phone-call-assistant-duplex-ethical-social-implications and they have a quote from Joanna Bryson, who talked at SAGE just before Christmas. Her quote is finished out with the following. There are no obvious answers to these questions, but as Bryson points out, Google is at least doing the world a service by bringing attention to this technology. It’s not the only company developing these services, and it certainly won’t be the only one to use them. ... (more)

AI and discovery tools

I got asked to contribute an answer to an upcoming feature looking at trends in scholarly discovery. There are four questions and I’ve been asked to think about an answer to the last of these four. In a sentence or two, please tell us what you understand is meant by ‘discovery’? For a researcher, how has the discovery experience developed in recent years? How can researchers decide which tools are going to do the right job for them? ... (more)

The robots are coming, the promise and perils of AI - questions

I’m at the Charleston conference, my first time, and we had a panel discussion this morning talking about AI. On the panel were: Heather Staines Director of Partnerships, Hypothes.is Peter Brantley Director of Online Strategy, UC Davis Elizabeth Caley Chief of Staff, Meta, Chan Zuckerberg Initiative Ruth Pickering Co-founder and Chief Strategy Officer, Yewno and myself. It was a pleasure to be on a panel with these amazing people. ... (more)