Top all-time · #932
Yann LeCun
Personal blog · yann.lecun.comaiscience
Yann LeCun's personal blog at yann.lecun.com covers ai and science. Since 2007, 74 different Hacker News readers have linked it 79 times, most often “The MNIST database of handwritten digits”. No feed was found, so read it on the site itself.
Yann LeCun's Home Page
Most-linked posts?
- The MNIST database of handwritten digits18 people · 20 mentions · first linked 2010-07-04
- A New Publishing Model in Computer Science 7 people · 7 mentions · first linked 2010-12-13
- Optimal brain damage4 people · 4 mentions · first linked 2011-05-20
- /exdb/publis/pdf/lecun-01a.pdf4 people · 4 mentions · first linked 2016-12-14
- /exdb/publis/pdf/bengio-lecun-07.pdf3 people · 3 mentions · first linked 2015-12-04
- Hardware Accelerated Convolutional Neural Networks for Synthetic Vision Systems [pdf]3 people · 3 mentions · first linked 2015-01-03
- /3 people · 3 mentions · first linked 2011-06-03
- Convolutional Neural Networks Machine Learning3 people · 3 mentions · first linked 2016-10-18
- A Theoretical Framework for Back-Propagation (1988) [pdf]3 people · 3 mentions · first linked 2017-06-27
- Fun Stuff (Yann LeCun)3 people · 3 mentions · first linked 2015-03-20
What Hacker News says?
I don't know where that "general consensus" came from, because in 1998 Lecun clearly showed [1] convnets beating all other models, including SVMs, at image recognition. [1] http://yann.lecun.com/exdb/publis/pdf/lecun-98.pdf
Comment linking /exdb/publis/pdf/lecun-98.pdf
on “Machine Learning: The Great Stagnation”Submitted A New Publishing Model in Computer Science as a story · 48 points
HN discussionThe first convnet that is similar to what we use today was described in 1989 [1], and immediately became the best method to do a handwritten digit recognition. [1] http://yann.lecun.com/exdb/publis/pdf/lecun-89e.pdf
Comment linking /exdb/publis/pdf/lecun-89e.pdf
on “Machine Learning: The Great Stagnation”Submitted "Steep Learning Curve" is an Erroneous Metaphor as a story · 43 points
HN discussionYea, so at Lambda, depending on the application, we use either convolutional neural networks or layered neural networks initialized with unsupervised training using autoencoders to do the feature extraction from images. The features are either passed through a softmax layer or used as a "siamese" network. (Siamese networks provide a similarity metric as opposed to a probability distribution over labels.[3]) As far a…
Comment linking /exdb/publis/pdf/chopra-05.pdf
on “Google Glass sales halted”Submitted A Theoretical Framework for Back-Propagation (1988) [pdf] as a story · 41 points
HN discussionThis is still a long way away from truly understanding the representational power of deep nets, but it's the first general progress that has been made since the early 90's. Exactly. Someone should tell Yann LeCun this --see e.g. Section 3.2 in [1], or pretty much every time he brings up circuit complexity theory to automagically imply that "most functions representable compactly with a deep architecture would requir…
Comment linking /exdb/publis/pdf/bengio-lecun-07.pdf
on “Machine learning works spectacularly well, but mathematicians aren’t sure why”Submitted A proposal for a new publishing model in Computer Science as a story · 31 points
HN discussion> Why is most AI work done in Lisp? Yann LeCun developed Lush, which is a Lisp for neural networks, during the early days of deep architectures. See https://yann.lecun.com/ex/downloads/index.html and https://lush.sourceforge.net . Things moved to Python after a brief period when Lua was also a serious contender. LeCun is not pleased with Python. I can't find his comments now, but he thinks Python is not an ideal sol…
Comment linking /ex/downloads/index.html
on “Hofstadter on Lisp (1983)”Submitted Optimal brain damage as a story · 16 points
HN discussionI went to a talk by Yann LeCun [1] a few months back, "Learning Feature Hierarchies for Vision." The current state of the art in this field is mind-blowing to an outsider. The final demo was a program that he trained in a matter of seconds to recognize and distinguish faces of various random audience members, in real time, from different viewing angles. [1] http://yann.lecun.com/
Comment linking /
on “How Perceptual Hashes Work”Submitted Proposal for a New Publishing Model in Computer Science (2009) as a story · 10 points
HN discussionThe format for the dataset is here: http://yann.lecun.com/exdb/mnist/ A good exercise would be to figure out how to extract the data and put it into a numpy array. Then you can test on most - if not all - of the frameworks.
Comment linking /exdb/mnist/
on “TensorFlow – Consise Examples for Beginners”Submitted Convolutional Neural Networks Machine Learning as a story · 5 points
HN discussionerr.. I think it is an evolution. Deep architectures allow for a more efficient function approximation from fewer examples than shallow architectures do. Deep (neural) networks are really just a generalization of machine learning (on graphs). The key is that we learn the similarity function to discriminate from examples instead of specifying it a priori. You can also build classifiers other ways: linear/logistic reg…
Comment linking /exdb/publis/pdf/bengio-lecun-07.pdf
on “An Introduction to Support Vector Machines”Submitted A New Publishing Model in Computer Science (2009) as a story · 3 points
HN discussionI think Norvig is a smart person, I enjoy his books, papers and jupyter notebooks, but I always thought he was pretty clueless regarding AI, as history indeed demonstrated. That's not an ad hominem. What is shameful was the extreme shunning of the mainstream scientific community to ANNs, in 90's-00's decades, to the point that it was considered career suicide to publish a ANN paper. I believe Yann LeCun has said sim…
Comment linking /ex/pamphlets/publishing-models.html
on “Yann LeCun on GPT-3”Submitted High Quality Document Image Compression with DjVu [pdf] as a story · 3 points
HN discussionWhich paper are you speaking about? Certainly, I'm always interested in a more complete history. I'm currently on LeCun's page and can't figure out which paper you're speaking to: http://yann.lecun.com/exdb/publis/index.html More generally, a common trope in NN papers and books is to draw a graph for matrix-vector multiplication and then draw the analogy that these are like neurons in the brain and this represents t…
Comment linking /exdb/publis/index.html
on “Coding the History of Deep Learning”Submitted The Psychology of Miles per Gallons as a story · 2 points
HN discussionChris, glad you like it. I can relate to that feeling, which I also had as an undergrad. It was a big motivation for this work -- and I think the analogy can go even further. For example, there's a really cool lecun paper about NN training dynamics actually being a lagrangian system: http://yann.lecun.com/exdb/publis/pdf/lecun-88.pdf . Another thing I want to try is literally write down a learning problem where S (t…
Comment linking /exdb/publis/pdf/lecun-88.pdf
on “Lagrangian Neural Networks”Submitted Fun Stuff (Yann LeCun) as a story · 2 points
HN discussionThe University of Toronto is home to Geoffrey Hinton and his research lab. Hinton was the first to use backpropagation for machine learning and a lot of the techniques used today were pioneered by his lab. He is also manages the Google Brain team in Toronto. Yann LeCun, the director of AI at Facebook, also did a postdoc at Toronto. [0] http://learning.cs.toronto.edu:40292/people#Faculty [1] http://yann.lecun.com/
Comment linking /
on “Alphabet’s Sidewalk Labs Eyes Toronto for Its Digital City”Submitted Fun Stuff for Data Scientists as a story · 2 points
HN discussionI agree (except for the first word), however I read the question with emphasis on "usual", as in, "What makes DNNs special?" There's pure performance (ex., in a Kaggle competition [ http://blog.kaggle.com/2012/11/01/deep-learning-how-i-did-it... ] or on a standard data set [ http://yann.lecun.com/exdb/mnist/ ], [ http://blogs.microsoft.com/next/2015/12/10/microsoft-researc... ] ), but that's what makes any ML method…
Comment linking /exdb/mnist/
on “What is the difference between deep learning and usual machine learning?”Submitted Hardware Accelerated Convolutional Neural Networks for Synthetic Vision Systems [pdf] as a story · 2 points
HN discussionThe way we do things in AI today is multiplication of two large matrices. Just like we did it 30 years ago: http://yann.lecun.com/exdb/publis/pdf/lecun-89e.pdf
Comment linking /exdb/publis/pdf/lecun-89e.pdf
on “Cerebras Systems unveils a 1.2T transistor chip for AI”Submitted Geoff Hinton Facts as a story · 2 points
HN discussion99.3% accuracy is not that great. Currently the best performer on this data set has 99.77% accuracy: http://yann.lecun.com/exdb/mnist/
Comment linking /exdb/mnist/
on “Tensorflow and Deep Learning, Without a PhD, Martin Gorner, Google [video]”Submitted The MNIST DATABASE of handwritten digits as a story · 1 points
HN discussion
