top | item 18874196

NLP Architect by Intel AI Lab

77 points| tsaprailis | 7 years ago |nlp_architect.nervanasys.com

11 comments

order

jph|7 years ago

This AI toolkit works on popular Intel CPUs, and is a big step forward for the new Intel Nervana Neural Network Processor (NNP-I) hardware chip akin to a GPU.

The Intel AI Lab has an introduction to NLP (https://ai.intel.com/deep-learning-foundations-to-enable-nat...) and optimized Tensorflow (https://ai.intel.com/tensorflow/)

One surprising research result for this NLP is that a simple convolutional architecture outperforms canonical recurrent networks, often. See: CMU lab, Sequence Modeling Benchmarks and Temporal Convolutional Networks (TCN) https://github.com/locuslab/TCN

If you're interested in Nervana, here are some specifics: the chip is for hardware neural network acceleration, for inference-based workloads. Notable features include fixed-point math, Ice Lake cores, 10-nanometer fabs, on-chip memory management by software directly, and hardware-optimized inter-chip parallelism.

I've worked for Intel, and I'm stoked to see the AI NLP progress.

modx07|7 years ago

By the way, the last author of the TCN paper (Vladlen Koltun) works at Intel Labs (Intelligent Systems Lab).

azinman2|7 years ago

So does this not work if you don’t have fancy new intel hardware?

continuations|7 years ago

How does this compare to word2vec or fasttext?

yorwba|7 years ago

word2vec and fasttext are specialized tools for creating word embeddings, this is a more generalist library. It's more comparable to PyText, AllenNLP or Flair, the main difference appearing to be that the other three use PyTorch, not Tensorflow.

___cs____|7 years ago

Yet another interface on top of Pytorch/TF/Gensim.