brain of mat kelcey


objax on honeysuckle farm

August 30, 2020 at 02:45 PM | categories: talk

recorded a high level explainer on objax while doing some chores. i think it's going to be a growing genre :)...

the map interpretation of attention

August 19, 2020 at 10:30 AM | categories: talk, three_strikes_rule

based on my three strikes rule i wrote a talk on the map interpretation of neural attention.it starts with the first NLP problem i saw where attention was used, goes through the way i think about attention in...

self supervised learning and making use of unlabelled data

July 02, 2020 at 05:00 PM | categories: talk

awhile back i did a talk on self supervised learning; here's a rerecording of it!...

an illustrative einsum example

May 27, 2020 at 12:00 AM | categories: talk

recently i changed what was quite clumsy looking code in something more elegant using a cool function in numpy called einsum and felt it was worth writing up.this example is just going to be...

using cross entropy for metric learning

May 19, 2020 at 06:00 PM | categories: talk

i gave a talk at the melbourne ml/ai meetup on using cross entropy for metric learning.here are the slides slidesand here's a recording; check it out!...

data engineering concerns for machine learning products

September 26, 2019 at 06:00 PM | categories: talk

i gave a talk at the melbourne data engineering meetup on "data engineering concerns for machine learning products"there was no recording :( (sad panda) but there are at least slides :)check them out!...

brutally short introduction to learning to learn

August 07, 2019 at 12:00 AM | categories: talk

i gave a talk at yow! data recently on learning to learn.there's an official recording on the yow datayoutube channel but since they missed out a bunch of my slides i did a slides-only-remix...

deep reinforcement learning for robotics

September 21, 2017 at 07:12 PM | categories: talk

did a talk on deep rl for robots at melbourne ml/aicheck it out!...

popular posts...

bnn : counting bees with a rasp pi


drivebot : learning to do laps with reinforcement learning and neural nets


wikipedia philosophy : do all first links on wikipedia lead to philosophy?


cartpole++ : deep RL hacking with a complex 3d cart pole environment


malmomo : deep RL hacking on minecraft with malmo


some papers from my time at google research / brain...

my honours thesis

the co-evolution of cooperative behaviour (1997) evolving neural nets with genetic algorithms for communication problems.

old projects...