Nevronsko strojno prevajanje

author: Matjaž Rihtar, Laboratorij za umetno inteligenco, Institut "Jožef Stefan"
published: Nov. 27, 2018,   recorded: November 2018,   views: 849
released under terms of: Creative Commons Attribution (CC-BY)


Related Open Educational Resources

Related content

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.
Lecture popularity: You need to login to cast your vote.


Predavanje ‘Nevronsko strojno prevajanje’ bo prikazalo trenutno stanje in izzive na področju nevronskega prevajanja. Prikazali bomo kratek uvod v nevronske mreže (nelinearnost, vzvratno razširjanje napak, možne izboljšave), nevronske jezikovne in prevajalne modele (LSTM, GRU, kodirnik-odkodirnik z mehanizmom pozornosti), možne izboljšave celotnega procesa (BPE, uporaba NMT pri prevajanju ne-naučenih jezikov) in trenutne izzive. Na koncu bomo v živo prikazali tudi prevajanje iz slovenščine v angleščino in obratno z nevronskim modelom, razvitim v Laboratoriju za umetno inteligenco na Institutu Jožef Stefan.

Link this page

Would you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !

Reviews and comments:

Comment1 jordan belfort, December 19, 2018 at 7:21 a.m.:

Then again, NMT models utilize intermittent neural systems. As per Läubli, "the beneficial thing here is that we can condition the likelihood of words that are created at each situation on all the past words in that yield sentence." generally, where SMT is constrained to what number of words its N-gram demonstrate directs, NMT assesses familiarity for the whole sentence.

Comment2 maxpetis, February 7, 2019 at 8:06 a.m.:

It is not the origin of this way of thinking, it's probably the beginning of him thinking of it like this. If one comes from a purely statistical or a probabilistic <a href="">prime assignment</a> learning background, suggesting things like this seems novel and fascinating. However, this is basically how you look at things in information theory.

Comment3 Law Assignment Writing, December 11, 2019 at 7:06 a.m.:

Treat assignment help offers custom assignment online where you can get excellent writing services from our Professional writers for your assignments. Treat Assignment Help is top-notch solution for custom <a href="">science-assignment-help</a> online in UK. Hire best coursework writers online and send your assignment requirements, our writing experts will give you affordable pricing quotation. for more info please email at: ! Call us at Toll Free Number Or Whatsapp at: +44 7520644027 ..!!

Write your own review or comment:

make sure you have javascript enabled or clear this field: