/100-research-papers

This lists all the research papers that I have read (and hopefully understood!)

MIT LicenseMIT

This repo will list my notes on all the research papers that I have read

(...and hopefully understood!)

Why only 100 papers! I don't know? I am not a researcher...in the traditional terms. But I do like to keep myself abreast of the latest developments in the ML/DL space. My interest is in its application. Its like what Andrew Ng has said - The power of AI/ML is not what SOTA models they build. But how they help in improving the lives of people.

This is my humble endevour to understand the latest research.

  1. A Large-scale benchmark for instance-level recognition and retrieval: This paper introduces the GLDv2 dataset Link

  2. M4 Competition- 100,000 time series and 61 forecasting methods: introduces the M4 competition dataset Link

  3. A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting: This article explains the method used by the M4 winning algorithm Link

  4. Analyzing Covid19 twitter stream: It's a summary of the research papers published using the Covid19 tweet dataset Link

  5. Analyzing three SOTA papers on NLP - first published in 2015