I'm starting a blog!

I’ve wanted to start a blog for a while. Now that I’m finally starting my PhD, it seems like the perfect time! So here it is, welcome to yet-another ML blog.

My field of study is natural language processing, and I’m looking at ways to exploit “external” language/knowledge more effectively in neural NLP systems. Most of my previous experience has been in QA systems, so initially this might dominate my posts.

What are you writing about?

My main aim is to post every two weeks, as a digest of what I’ve been thinking about and reading about. I’m also hoping to post the odd bit of news and generally to have a platform to publish on, but most of the time, posts may be quite long, quite dry and quite serious… I’ve probably lost you already.

I also hope that by sharing my thoughts and progress, it might be of some use to others in the field! I think a fair number of the posts will consist of paper digests and groups of related papers on a particular theme. These might be helpful to you as a good starting point to introduce a sub-field you might be unfamiliar with, or to see a sub-field from a new perspective.

Why are you writing it?

First and foremost, I’m hoping to use the blog as a organisational tool. The idea is that by working towards a post every couple of weeks, I’ll be able to structure and organise my reserach more effectively, and be more efficient.

It will also give me something to focus on when there are inevitable difficult and slow periods in my PhD journey.

In addition. I’m hoping that having a neat, easy-to-share record of what I’m up to will make it easier and faster to get my ideas out to the community and get feedback (anything from criticism, paper recommendations, bits of advice or Q and A).

So, whilst the blog is mostly for me, I hope that you’ll engage with it, and you’ll get something out of it too 😊

When will you post?

I’m aiming for every two weeks. We’ll see how this goes 🙃

My writing won’t get better without your feedback, so please do let me know what you think!

comments powered by Disqus