akssri

This is probably the only project which goes about building a backplate from household tools, rather than having it milled.

It should be noted that things like momentum etc in DNN training are inspired directly from Nesterov, and apparently work very well.

It's probably (n^3 * n^2) vs (n^2 * n^2) , and not cubic vs linear as was stated, where (n = 1/h).

Nesterov's method attains super-linear convergence on Convex functions. They don't really (if I remember right) work otherwise; but then this is okay, since Convex problems are really the only ones we can solve in poly-time (esp. if P/=NP and UGC is true).

SA/RR are global optimization methods which come with very little (if any?) complexity guarantees.

AlJazeera has a nice documentary, https://www.youtube.com/watch?v=8OsMdzrxQfA

It appears that tusks from African elephants are more valued in China than those from S/S.E Asia. Does anyone know why this is ?

Ah, is this https://github.com/CNugteren/CLBlast ?

I was looking at, https://github.com/uncomplicate/neanderthal/blob/master/src/...

Ah, you're right; 5.6 Gflops is for Hawaii. That's doubly impressive. I'll make sure to try your kernel then. Thank you!

I think it's about 5.6 Tflops. Wow, 3.75 Tflops on Hawaii is very good indeed; I agree this is not something that clBLAS would beat by a wide-margin if at all.

I found the latest version of clBLAS on Fiji achieves a fantastic ~4 Tflops (on 2^n matrices). NVblas has probably had more resources allocated to it that clBLAS. I'd be positively surprised if the kernels in Neantherdal beat those. Do you plan on adding benchmarks for the GPU calls ? I can help running the clBLAS benchmarks, if you like, since I have a tuned setup.

If you want to take a look at it, the AutoGemm generator seems to be a simple python script written in order to overcome the limitations of the C preprocessor. I was considering using its tiling structure, since I already have a Lisp->OpenCL compiler in place (and have had no luck beating it). See, for instance,

https://github.com/matlisp/matlisp-opencl/blob/master/tests/...

https://github.com/matlisp/matlisp-opencl/blob/master/src/te...

Sweet! Did you consider using AutoGEMM.py from clBLAS instead of a static GEMM kernel ?

I was considering using a polyhedral compiler macro (like PPCG), for writing OpenCL kernels in matlisp, but it's not clear how optimal this would be.

This of course was political, much as it is today in Tibet.

As an aside, I've always found the way in which East Asian Buddhists held firm against cultural invaders, extremely inspiring. In India, there seems to be little other than memories of conspiring with the colonizers for profit.

I think he meant Protestantism in particular.

One wonders if this on a meta-philosophical level is related to the acceptance of death in Eastern cultures as just the beginning of another birth. Or may be it's a clever device to make people become attached to life ? Hmm.

Actually there is one such living practice amongst Jains, https://en.wikipedia.org/wiki/Santhara

sAkyamuni himself did, and many sAdhus today go to great trouble to become "asaktaH" (non-attached). It's said that former recognized the errors in this path, gave up on this course, and attained nirvana the subsequent day. Unless this is modern propaganda, I imagine Kobo Daishi was aware of it too.

India too has not suffered a loss of creatures who file PILs at the slightest deviation from the fashionable.

(note: Jainism is a traditionally recognised nAstika school, one that's not grounded in the vedAni. By this definition, Buddhism too is Hinduism - the demarcation only became rigid in the past ~200 years.).

Rather odd that the course omits decision trees.

Perhaps this will help ? https://akssri.github.io/blog/Kalman-Filtering.html

These algorithms can't really make much use of Tensorflow.

If you can call those accents "American" (subject of 2+ such scam calls here).

> "'We have a new theorem--that mathematicians can only prove trivial theorems, because every theorem that's proved is trivial.'" - Richard Feynman

Sigh.

KF is essentially solving a QP with equality constraints (Boyd's course is a good place for details), which can be solved exactly with a single decomposition of the KKT system - picking an ordering is all that matters for complexity.

This is essentially the principle on which all of Sparse Linear algebra and Graphical models work. There is nothing special about the structure of KF, nor in LQR, nor in their non-linear generalizations.

One can symbolically unroll Schur complements multiple times to make block-LU appear opaque and sophisticated, but it really is not (this of course is not to say this is done deliberately). KF can also derived from the Bayes' network model, but extending this to non-linear forms like EKF, and to things which are not first-order becomes rather troublesome (or impossible).

I'd have appreciated a post asking for details rather than infantile derision.

KF is essentially a symbolic unrolling of a LU decomposition. There is a graphical algorithm which makes all the minutae trivial [1].

[1] https://arxiv.org/pdf/1508.00952.pdf

Fun fact: Symbolics continues to sell 'Macsyma'.

I think it's also important to note that Macsyma (or atleast Maxima) has a lot of cruft in the code. I actually like Weyl better, even if it hasn't seen active development in ~2 decades.

P.S: Maxima actually uses a list-of-lists to represent a matrix.

What exactly were these 'problems' ?

Exactly. I think the libertarian way to go would be to limit the role of the Govt. to enable Tax breaks for spouses, and other impersonal things.

> but we value choice and freedom over rules and processes.

Which I exactly why I feel Lisp doesn't see much use elsewhere :(

Yeah it's fairly easy to fix this (and you can even do it without unwinding the stack!). There are tons of sbcl global variables which you can tweak on-the-fly.

> but for many, its taste cannot compare to the smooth and complex flavours of Arabica.

The author obviously hasn't had an awesome cup of Filter Kaapi :)

http://en.wikipedia.org/wiki/Indian_filter_coffee