aisofteng

After reading the links you gave, I’d like to clarify that I meant that edge cases that can cause data loss are not common enough, in my experience, for one to be forced to treat ElasticSearch as an unreliable data store. Anecdotally, I’ve never seen it happen. You’re right, maybe I’m lucky; I would like to see a case study of some project where these issues were significant enough to treat ElasticSearch as an unreliable data store.

These are standard information retrieval algorithms that have been known and used for decades.

As someone who has used ElasticSearch in production for years, as far as I know this is simply not true.

Solr and Elasticsearch are not relational database systems and have none of the optimizations for relational data that Postgres does. They are built on top of Lucene and target document retrieval via an inverted index [1] rather than dealing with relational data.

[1] https://en.wikipedia.org/wiki/Inverted_index

“The whole wolpertinger thing” is a metaphor and a literary device. This isn’t a technical manual.

I don’t know why you found this hard to read. The writing is clear and understandable. The dismissiveness of your comment and the fact that its out-of-handedness is based on nothing objective suggests that you’re not the target audience, especially since, in contrast to your comment, the thoughts in this article are researched, decently sophisticated, and exposed in a discursive manner.

Any engineer that thinks metaphysics is “b.s.” is uneducated in the schools of thought that made engineering of any sort possible in the first place.

>I found myself more willing to consider that the concrete and immediate expression of some piece of program logic might serve better than trying to pull out a more generic version of the same logic that might never be re-used in practice.

This is the YAGNI principle in XP - “you ain’t gonna need it”: https://en.wikipedia.org/wiki/You_aren%27t_gonna_need_it

Elasticsearch won’t do that out of the box either. How would you solve this in the general case without causing false positives?

SQLite with Raft. Nice project.

No, it doesn’t. A trait that is not deleterious to reproduction will not be naturally selected against.

Would Tini help?

https://github.com/krallin/tini/blob/master/README.md

I had courses in each of those topics as part of my undergraduate math degree. There’s no reason an AI degree couldn’t include those as standard curriculum for undergraduates.

For more on what's different after the octonions, see: https://en.m.wikipedia.org/wiki/Hurwitz%27s_theorem_(composi...

It isn't explicitly stated there, but the statement I recall from algebraic topology is that the octonions are the last normes division algebra.

>Teaching essentially involves opening each image, converting it into a base64 string, and submitting it to the /classificationbox/teach API endpoint in a request that resembles this.

Why base64?

No, IBM has not done that. IBM Global Business Services and IBM Global Technical are their own business units.

I generally understand the title of senior engineer at an organization as indicating that the so titled person can be relied on to solve the majority of engineering needs of that organization - essentially, a senior engineer is the person you ask how to do something when you don't know how to do something.

>Another problem of word vectors is that any word might actually have multiple senses, while vectors are just point estimates. If we wanted to be correct, we needed first to find the right sense for each word in a phrase and only then assign the vector. There is research in "on-the-fly" word vectors that adapt to context, but it's much harder to use.

There has been work on representing words not as vectors but as multimodal Gaussian distributions in order to try to deal with polysemy, such as [0], of which an implementation (which I have not tried to use) is available on GitHub at [1].

>It feels different than 3d-space because each point has a shortcut to other points, each point leads to hundreds of other places which might be far apart. It's like a kaleidoscope where a small change can create a very different perspective.

I appreciate that the author of the parent comment has found some sort of intuition, but I would caution others from trying to use the above quote in order to develop their own intuition as it is meaningless in any rigorous sense.

[0] https://arxiv.org/abs/1704.08424

[1] https://github.com/benathi/word2gm

>FastText [0] in particular is quite impressive. It is essentially a variant of Word2Vec that also supports n-grams,

To clarify: fasttext does not support n-grams of words, but instead considers n-grams of characters within words.

OpenWhisk is what runs IBM Cloud Functions: https://console.bluemix.net/openwhisk/

That's wage discrimination and is illegal.

Neither of you has said explicitly what you mean by "better" and likely have different measures in mind when saying it here.

>Linear programming (also called linear optimization) is an applied field of mathematics often used in operations research and planning. It attempts to find an optimal solution to a planning problem when a set of business constraints exist.

No, linear programming is an applied field of mathematics that attempts to find an optimal solution to a goal within a system which is under some sort of constraint(s). Whether those constraints are business constraints is irrelevant; the theory exists outside of any particular application.

The point is that to produce content requires investment and/or valuable skill. You've clearly never had a livelihood that depends on knowledge, experience, or skills that took you years to develop and then has someone copy it and profit from it. That's what copyright is for.

>There is definitely a learning curve to state machines,

In my view, the concept of a finite state machine is a simple one with essentially no learning curve. It is a straightforward concept. I would never hire anyone that described experiencing a learning curve for this concept.

It's disappointing that this needs to be spelled out. It really should be common sense, minimal competency.

A tangential comment, but - what excellent styles of both writing and exposition!

In particular, PageRank can't be used for corpora whose documents do not link to each other explicitly somehow - which, outside of hypermedia, is nowhere.

I found it quite straightforward. There isn't any "jumping around".

Essentially blogspam.

https://lkml.org/lkml/2018/1/3/797 is the original source.

Well, it isn't their fault.