davnn

14 ms·

121.
by davnn 6y ago
Non-paywalled version: https://archive.is/VKdqx
122.
by davnn 6y ago
So the main claim of this paper is: "...good models and training procedures are those that (1) optimize quickly in the ideal world and (2) do not optimize too quickly in the real world." Will be interesting to replicate those resu
123.
by davnn 6y ago
Absolutely! I'm just not sure if we have chosen the right notation to describe our ideas. There are many interesting developments in formalizing mathematics and I hope that the field of mathematics is gravitating towards such solutions
124.
by davnn 6y ago
Well, I believe the authors do their best to convey their ideas, but due to the fact that the notation is mostly informal it's easy to miss details. Also, different institutes might use their own special notation to describe the same i
125.
by davnn 6y ago
More of a rant, but since I began to implement algorithms from various papers for my PhD, my aversion to mathematical notation grows steadily. More often than not it's imprecise and hard to reproduce.
126.
by davnn 6y ago
I can't imagine that prison life is so disconnected from the rest of society; it certainly isn't in Europe.
127.
BMW i4 unveiled with 590 km (WLTP) and up to 300 miles (EPA) range(bmwblog.com)
1 points by davnn 6y ago|2 comments
128.
by davnn 6y ago
I would not see Zip as a file format, it's just compressed version of some other format. You therefore inherit all the pros and benefits of the compressed file, which might be columnar storage (e.g. Parquet) or row-based (e.g. Avro), b
129.
by davnn 6y ago
Does anyone have insights into why data.table is as fast as it is? PS: Great work with polars!
130.
by davnn 6y ago
The intersection is also interesting when you look at equational proofs in Mathematica. https://reference.wolfram.com/language/ref/FindEquationalPro...
131.
by davnn 6y ago
Relating to other programming languages, it was the first term that came to mind. You have to know the type before you can specialize a function, don't you? I think it's a good mental model to constantly keep the types in mind, bu
132.
by davnn 6y ago
That seems to work really well. I didn't have use cases where I would prefer to call Python from Julia instead of using Python directly. Additionally, I think you also have to think about things like Documentation and Tooling.
133.
by davnn 6y ago
First and foremost you should have a general understanding of type inference. You also have to understand the difference between a function and a method (in Julia's terms), see: Functions: https://docs.julialang.org/en&
134.
by davnn 6y ago
Really interesting work! I'm really happy that people at MIT are pushing forward Julia. In my opinion, Julia is incredibly powerful once you understand the language. It often feels like writing pseudocode that performs like C++. The ec
135.
by davnn 6y ago
2016 for v1 to be exact, link: https://arxiv.org/abs/1611.03530 and discussion: https://news.ycombinator.com/item?id=13566917
136.
by davnn 6y ago
On the flip side: how to stay focused in a loud office? It's surprisingly difficult for me compared to working from home. Edit: Suggestions for OP. Find purpose; keep your workplace tidy and if possible try to separate work from your f
137.
by davnn 6y ago
Could be adblocked as well.
138.
by davnn 6y ago
> BTW, I am a huge fan of your writing: the Principle of Least Power is a sacred programming text to me . That principle is actually directly from the CTM book, which seems to have been a great inspiration for Martin Odersky in the langu
139.
by davnn 6y ago
Interesting idea! Is there any research in this direction?
140.
by davnn 6y ago
The fun thing is that there are approaches that want to go beyond this kind of methodological description of a scientific process to code [0, 1]. In general I would say that the more we can remove the human aspect and inherent ambiguity
141.
by davnn 6y ago
> Re-using source code saves time, but I would argue not being able to shouldn't threaten reproducibility. More often than not it‘s not clear from a paper what exactly the authors did to a achieve a specific result. Being able to ex
142.
by davnn 6y ago
> Discussions about the efficiency/environmental impact/... of combustion v.s. electric aside, it's obvious that driving a huge battery around will burn more fuel. I am not sure if I am misunderstanding something here, but
143.
by davnn 6y ago
Just look at the page source for CSS and JS used to create it.
144.
by davnn 6y ago
Have a look at the Stanford Machine Learning Group projects [1]. What most articles miss in my opinion is that AI progress heavily depends in the data being available and that‘s where there are huge opportunities in the future. [1] https:
145.
by davnn 6y ago
> It may well be that this applies in individual cases. Well from a machine learning perspective it will probably be in all cases where enough, qualitative data is available and not a lot of contextual knowledge is necessary that cannot
146.
by davnn 6y ago
What problems did you experience in web development on Windows? I worked with Linux, Mac and Windows, but I didn‘t have any problems on any platform with a typical modern webpack/react/angular/typescript/elm etc.. stack.
147.
by davnn 6y ago
There is also the abstract algebra cheatsheet [1]. Not my work, I have just bookmarked it a couple of years ago. [1] https://github.com/mavam/abstract-algebra-cheatsheet
148.
by davnn 7y ago
I was mainly referring to CT scan data. A symptomatic check should not require much experience, rather applying the knowledge in form of a simple decision tree. Regarding CT scans .. machine learning will be so much better than humans in cl
149.
by davnn 7y ago
> This also why I believe human doctors are still unbeatable by AI. Because you hypothesize that they can rapidly learn new, not very distinctive patterns without much data?
150.
by davnn 7y ago
So the difference between this an google‘s reverse image search is that google‘s matching algorithm is worse (probably because it doesn‘t include facial recognition)? Well, public images are public and I don‘t think banning such a service w

More ›