ALBERT: A Lite BERT for Self-Supervised Learning of Language Representations
http://bit.ly/2SOUCmC
Microsoft Research Open Data Project: Evolving our standards for data access and reproducible research
http://bit.ly/2LQMr4L
eBay’s Transformation to a Modern AI Platform
http://bit.ly/2sH7IYg
Beyond L2 Loss — How we experiment with loss functions at Lyft
https://lft.to/2rTywEm
subscribe @datafolks
http://bit.ly/2SOUCmC
Microsoft Research Open Data Project: Evolving our standards for data access and reproducible research
http://bit.ly/2LQMr4L
eBay’s Transformation to a Modern AI Platform
http://bit.ly/2sH7IYg
Beyond L2 Loss — How we experiment with loss functions at Lyft
https://lft.to/2rTywEm
subscribe @datafolks
Google AI Blog
ALBERT: A Lite BERT for Self-Supervised Learning of Language Representations
Posted by Radu Soricut and Zhenzhong Lan, Research Scientists, Google Research Ever since the advent of BERT a year ago, natural languag...
Best Data Visualization Projects of 2019
http://bit.ly/36drg4e
5 Weird and Hilarious Uses of Data Science
http://bit.ly/37pq4fi
subscribe @datafolks
http://bit.ly/36drg4e
5 Weird and Hilarious Uses of Data Science
http://bit.ly/37pq4fi
subscribe @datafolks
FlowingData
Best Data Visualization Projects of 2019
As I do every year, I picked my ten favorite visualization projects. Here they are in no particular order.
Plug and Play Language Models: A Simple Approach to Controlled Text Generation
https://ubr.to/37M4q5a
subscribe @datafolks
https://ubr.to/37M4q5a
subscribe @datafolks
Uber Engineering Blog
Plug and Play Language Models: A Simple Approach to Controlled Text Generation
S. Dathathri, A. Madotto, J. Lan, J. Hung, E. Frank, P. Molino, J. Yosinski, R. LiuLarge transformer-based language models (LMs) trained on huge text corpora have shown unparalleled generation capabilities. However, controlling attributes of the generated…
RoboNet: A Dataset for Large-Scale Multi-Robot Learning
https://stanford.io/2OSu1RX
subscribe @datafolks
https://stanford.io/2OSu1RX
subscribe @datafolks
ai.stanford.edu
RoboNet: A Dataset for Large-Scale Multi-Robot Learning
Towards a Conversational Agent that Can Chat About… Anything
http://bit.ly/2OfIKGS
subscribe @datafolks
http://bit.ly/2OfIKGS
subscribe @datafolks
Google AI Blog
Towards a Conversational Agent that Can Chat About…Anything
Posted by Daniel Adiwardana, Senior Research Engineer, and Thang Luong, Senior Research Scientist, Google Research, Brain Team Modern co...
Introducing PyTorch3D: An open-source library for 3D deep learning
http://bit.ly/2uqTz2i
subscribe @datafolks
http://bit.ly/2uqTz2i
subscribe @datafolks
Facebook
Introducing PyTorch3D: An open-source library for 3D deep learning
We just released PyTorch3D, a new toolkit for researchers and engineers that’s fast and modular for 3D deep learning research.
Plug and Play Language Models: A Simple Approach to Controlled Text Generation
https://ubr.to/37M4q5a
subscribe @datafolks
https://ubr.to/37M4q5a
subscribe @datafolks
Uber Engineering Blog
Plug and Play Language Models: A Simple Approach to Controlled Text Generation
S. Dathathri, A. Madotto, J. Lan, J. Hung, E. Frank, P. Molino, J. Yosinski, R. LiuLarge transformer-based language models (LMs) trained on huge text corpora have shown unparalleled generation capabilities. However, controlling attributes of the generated…
BERT, ELMo, & GPT-2: How Contextual are Contextualized Word Representations?
https://stanford.io/2RKUBOY
RoboNet: A Dataset for Large-Scale Multi-Robot Learning
https://stanford.io/2OSu1RX
subscribe @datafolks
https://stanford.io/2RKUBOY
RoboNet: A Dataset for Large-Scale Multi-Robot Learning
https://stanford.io/2OSu1RX
subscribe @datafolks
SAIL Blog
BERT, ELMo, & GPT-2: How Contextual are Contextualized Word Representations?
Geometric properties of contextualized word representations and what that says about how contextualized they are.
Code and Dataset for Indonesia Sentiment Analaysis Apps Review https://github.com/andreaschandra/google-play-review
GitHub
GitHub - andreaschandra/google-play-review: Sentiment Analysing to Indonesian Apps on Google Play
Sentiment Analysing to Indonesian Apps on Google Play - GitHub - andreaschandra/google-play-review: Sentiment Analysing to Indonesian Apps on Google Play
Our latest articles
Searching A Novel Method in ML Research
https://bit.ly/3cHBtKk
My Work on Speech Processing
https://bit.ly/2WHvcst
subscribe @datafolks
Searching A Novel Method in ML Research
https://bit.ly/3cHBtKk
My Work on Speech Processing
https://bit.ly/2WHvcst
subscribe @datafolks
Medium
Searching A Novel Method in ML Research
It started when I had an ambition to publish paper on a ranked journal which your paper (may) should propose a new method, like literally…
Our latest work on paraphrase identification has just released in arXiv!
A duplicate question is inevitable in a QnA forum. To identify this problem, we experimented on Paraphrase Identification using the Quora Question Pairs dataset. We achieved a stunning result by using several methods and algorithms such as LSTM, BERT, and different approach from the previous works to improve the model performance. Here is the full paper of how we experimented the Paraphrase Identification: arxiv.org/abs/2006.02648
A duplicate question is inevitable in a QnA forum. To identify this problem, we experimented on Paraphrase Identification using the Quora Question Pairs dataset. We achieved a stunning result by using several methods and algorithms such as LSTM, BERT, and different approach from the previous works to improve the model performance. Here is the full paper of how we experimented the Paraphrase Identification: arxiv.org/abs/2006.02648