Top all-time · #16822
Ankesh Anand
Personal blog · ankeshanand.com/blogaiscience
Ankesh Anand's personal blog at ankeshanand.com covers ai and science. Since 2014, 4 different Hacker News readers have linked it 7 times, most often “Contrastive Self-Supervised Learning”. The last post was in 2022, but the archive is still widely read.
Too few sampled comments (4) to say how often HN recommends it.?
Why HN links it
The blog's posts explain self-supervised learning, including a fully illustrated article on contrastive predictive coding. Commenters cite it as a reference for explaining CPC's classification approach.
Summarised by Claude from a sample of 4 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
PhD student in Artifical Intelligence at Mila, Montreal
Most-linked posts?
- Contrastive Self-Supervised Learning4 people · 4 mentions · first linked 2020-01-30
What Hacker News says?
A fully illustrated article [1] And Lilian Weng blog on self-supervision [2] .. CPC is .. translating a generative modeling problem to a classification problem... uses cross-entropy loss to measure how well the model can classify the “future” representation amongst a set of unrelated “negative” samples... [1] https://ankeshanand.com/blog/2020/01/26/contrative-self-supe... [2] https://lilianweng.github.io/lil-log/201…
Comment linking /blog/2020/01/26/contrative-self-supervised-learning.html
on “Visual Introduction to Self Supervised Learning”Submitted Contrastive Self-Supervised Learning as a story · 97 points
HN discussionI would modify the article to say "the next leap in AI is prediction". There is a lot of activity in this space called Predictive Coding. Look up the work of David Cox[1], Karl Friston et al. A fascinating variant is Contrastive Predictive Coding[2]. [1] https://www.youtube.com/watch?v=P0yVuoATjzs [2] https://ankeshanand.com/blog/2020/01/26/contrative-self-supe...
Comment linking /blog/2020/01/26/contrative-self-supervised-learning.html
on “The Next Leap in Self-Driving: Prediction”Submitted Reinforcement Learning as a fine-tuning paradigm as a story · 22 points
HN discussionLocation: Kharagpur, India Remote: No Willing to relocate: Yes (San Francisco / Bay area preferably) Technologies: Python, Javascript, Django, Flask, MySQL, MongoDB, Redis, HTML, CSS, ReactJS (Full stack); scikit-learn, Pandas, Numpy, NetworkX (Data science) Resume/CV: http://ankeshanand.com/CV.pdf Email: ankeshanand@iitkgp.ac.in
Comment linking /CV.pdf
on “Ask HN: Who wants to be hired? (August 2015)”Submitted Contrastive Self-Supervised Learning – Ankesh Anand as a story · 2 points
HN discussionLocation: West Bengal, India. Actively looking to relocate, Bay Area preferably. Remote: No Languages: Python, PHP, Javascript, HTML5, CSS3, C++, C, Matlab, MySQL, Assembly Framweorks / Libraries: Django, Flask, ROS, Qt, OpenCV, d3.js, Leaflet, Bootstrap - Google Summer of Code student for BRL-CAD in 2014. - Maths and CS undergrad at IIT Kharagpur. - Spent a summer researching at the Max Planck Institute of Software…
Comment linking /
on “Ask HN: Who wants to be hired? (October 2014)”