Humphrey Sheil


Differentiable Neural Computers (DNCs) - Nature article thoughts

17 Oct 2016 | 17 comments

The addition of working memory to artificial neural networks (ANNs) is an obvious upgrade when we compare ANNs to the Von Neumann CPU architecture, and one that came to the fore in the RAM (Reasoning, Attention, Memory) Continue reading  

Pre-trained embeddings into Torch LookupTables before LSTM training

13 Jul 2016 | 3 comments

A promising method to improve performance and reduce training time for an RNN / LSTM is to (a) use embeddings / LookupTables and (b) pre-train those embeddings before commencing full-blown training. I'm interested in how much better a...

Continue reading  

Powering the next-gen of deep learning: Pascal and CUDA 8

06 Apr 2016 | 2 comments

A lot of the time in Deep Learning, the GPU is abstracted away by whatever framework you happen to be using. Sure you run out of memory and drop down the batch size or some other parameter, or you look at nvidia-smi and wonder why you're not running...

Continue reading  

Getting to grips with LSTM (part one)

14 Jan 2016 | 26 comments

What is LSTM (Long Short Term Memory)

Have you got a data set which you are sure contains some really interesting temporal relationships? Do you suspect that if only you could exploit this knowledge you could improve your target predictor /...

Continue reading  

A quick run through Vowpal Wabbit

08 Jan 2016 | 1 comments

Vowpal Wabbit (abbreviated to VW from now on) came to my attention again at NIPS 2015 - I first saw it at NIPS 2013 and I've heard good things about it, so I decided to give it a test...

Continue reading  
1 2