charleshmartin

8 ms·

31.
by charleshmartin 7y ago
Available. Expertise in ruby, python, ML, AI, data science
32.
by charleshmartin 7y ago
I'm developing the weightwatcher tool for deep neural networks into a full fledged product http://github.com/CalculatedContent/WeightWatcher The weightwatcher lets you detect potential problems in a trained neural
33.
WeightWatcher: Empirical Quality Metrics for Deep Neural Networks(calculatedcontent.com)
2 points by charleshmartin 7y ago|0 comments
34.
Towards a New Theory of Learning: Statistical Mechanics of Deep Neural Networks(calculatedcontent.com)
5 points by charleshmartin 7y ago|0 comments
35.
by charleshmartin 7y ago
How it is done: https://aip.scitation.org/doi/10.1063/1.1906206
36.
by charleshmartin 8y ago
I would say “Understanding Deep Learning Requires REMEMBERING Generalization” https://calculatedcontent.com/2018/04/01/rethinking-or-remem... https://arxiv.org/abs/1710.09553 We can unde
37.
by charleshmartin 8y ago
An excellent paper which uses (some of the) results we have also found studying the weight matrices of neural networks...namely that they rarely undergo rank collapse https://calculatedcontent.com/2018/09/21/r
38.
by charleshmartin 10y ago
The paper is pretty simple. It just shows that Hinton's scheme for introducing hidden variables, and then minimizing, is a type of variational RG, introduced by Kadanoff 10 years earlier Beyond that, the analogy is not very strong, IM
39.
by charleshmartin 10y ago
See my blog: https://charlesmartin14.wordpress.com/2015/04/01/why-deep-le... I try to highlight the paper and some of the history and relevance, in a general way, but also hitting the math hard. the more gene
40.
by charleshmartin 10y ago
SEEKING WORK Location: San Francisco, CA Experts in machine learning, data science, and software development. NLP, Ad click prediction, anomaly detection, image and signal classification, deep learning. Java, Python, Ruby. I have worked wi
41.
by charleshmartin 10y ago
That is helpful thanks.
42.
by charleshmartin 10y ago
It is difficult to understand the implications of these assumptions and if they really apply to supervised deep learning nets. It has been know for a very long time that simple models, like the Random Energy Model, displays a spin glass tra
43.
by charleshmartin 11y ago
I will address the supervised vs unsupervised issue in my next post. Here, I believe the analogy would be that when a field is applied to a spin glass, it does not exhibit a glass transition to a non-self-averaging (highly non-convex) grou
44.
by charleshmartin 11y ago
It is easier to apply the Sornette theory to antibubbles. Bitcoin seemed like a great example. I gotta go back and see how well the predictions actually worked.
45.
by charleshmartin 11y ago
'Technically' this is correct--the RBM CD algo is not minimizing this function; that's not the point. It is known that when training an RBM, the reconstruction error decreases but not monotonically; in fact it fluctuates. In
46.
by charleshmartin 11y ago
Thanks for the comments. It is helpful to have others read the blog and make suggestions and ask for clarifications. The motivations here are to (1) summarize and clarify the key ideas of the physics paper that observed this connection
47.
by charleshmartin 11y ago
I am happy to answer questions. You can ask here or directly on the blog.