8 ms·
Pretty amazing that you can patent an idea as simple as multiplying a matrix by itself N times.
by umanwizard 7y ago
Pretty amazing that you can patent an idea as simple as multiplying a matrix by itself N times.
- shoo 7y agoapplied exponentiation
- enitihas 7y agoWell it seems much better when you compare it to patenting of rounded corners.
- ekianjo 7y agoWell some companies patented one click to order. Hard to beat.
- make3 7y agoperson, the question actually is which matrix that is
- IfOnlyYouKnew 7y agoConsidering there were search engines before Google, the idea apparently wasn’t quite as obvious.
- umanwizard 7y agoSimple does not necessarily imply obvious. The chain rule is another simple application of matrix multiplication that most people couldn’t have come up with independently.
- gilgoomesh 7y agoI realize that not everyone understands how patents work but this is ridiculous. The patent claims don't mention matrices. Any implementation (like matrices) is merely an embodiment (you can implement the patent without matrices). And even if that weren't true, the foundation of the patent system is applying existing techniques to new applications. The background section of the patent clearly details how this technique has been used in other applications. I don't dispute that a lot of patents are trash but this is possibly the most important patent of the last 30 years. That doesn't mean it invented computers, mathematics and the internet, it just put some already good ideas together. That is what invention is.
- Matumio 7y agoThe idea sure was brilliant at the time. But I really doubt that we (as a society) have gained anything by allowing this to be protected. It might have prevented some healthy competition. Certainly anyone thinking about search engines (once this became a thing) would have thought of this. And I doubt that nobody would have bothered to create a search engine just because it couldn't be patented.
- Certhas 7y agoI realize that not everyone understands how matrices work, but this is ridiculous. The abstract of the patent: "... the rank of a document is calculated from a constant representing the probability that a browser through the database will randomly jump to the document." And the specific claim is: "Looked at another way, the importance of a page is directly related to the steady-state probability that a random web surfer ends up at the page after following a large number of links." This _explicitly_ describes a Markov Chain, which is naturally represented by a matrix. A variety of versions of the linear equation are explicitly given in the patent. To claim you can implement the patent without matrices is, for all intents and purposes, wrong. You can implement the same equations in a variety of ways, but they are still matrix equations. They patented the idea to apply random walks to ranking webpages. That's arguably reasonably novel, though Wikipedia lists a number of predecessors. But it was also an inevitable invention, because there is a large number of people familiar with Random Walks/Markov Processes, they are routinely taught to undergrads, and are used to model and analyse a vast number of processes [1]. [1] https://en.wikipedia.org/wiki/Markov_chain#Applications https://en.wikipedia.org/wiki/Markov_chain#Applications
- sideshowb 7y agoIt's the application to search that was patented. There were plenty prior publications on eigenvalue centrality.
- mojuba 7y agoIt's how the US patent system works. Basically it's almost like copyright: you can patent virtually anything that hasn't been patented yet until challenged in court as either trivial or not original (i.e. prior art exists). Essentially the US patent system achieves two goals: shift the burden of verifications and validations to competitors and courts; and secondly, provide the US some advantage on the international level via the Patent Cooperation Treaty. Any crappy patent may turn out to be important, so just let them register everything and see which one "sticks".
- bhl 7y agoIf we’re talking about finding the steady / convergence state of a matrix, there are better ways than simple repeated matrix multiplication. One would be to solve a linear system of equations. The one PageRank uses if I recall correctly is the power method for eigenvalues.