|
Size: 834
Comment:
|
Size: 5810
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 3: | Line 3: |
| = Natural Language Processing Seminar 2016–2017 = | = Natural Language Processing Seminar 2020–2021 = |
| Line 5: | Line 5: |
| ||<style="border:0;padding:0">The NLP Seminar is organised by the [[http://nlp.ipipan.waw.pl/|Linguistic Engineering Group]] at the [[http://www.ipipan.waw.pl/en/|Institute of Computer Science]], [[http://www.pan.pl/index.php?newlang=english|Polish Academy of Sciences]] (ICS PAS). It takes place on (some) Mondays, normally at 10:15 am, in the seminar room of the ICS PAS (ul. Jana Kazimierza 5, Warszawa). ||<style="border:0;padding-left:30px">[[seminarium-archiwum|{{attachment:pl.png}}]]|| | ||<style="border:0;padding-bottom:10px">The NLP Seminar is organised by the [[http://nlp.ipipan.waw.pl/|Linguistic Engineering Group]] at the [[http://www.ipipan.waw.pl/en/|Institute of Computer Science]], [[http://www.pan.pl/index.php?newlang=english|Polish Academy of Sciences]] (ICS PAS). It takes place on (some) Mondays, normally at 10:15 am, in the seminar room of the ICS PAS (ul. Jana Kazimierza 5, Warszawa). All recorded talks are available on [[https://www.youtube.com/ipipan|YouTube]]. ||<style="border:0;padding-left:30px">[[seminarium|{{attachment:seminar-archive/pl.png}}]]|| |
| Line 7: | Line 7: |
| ||<style="border:0;padding-top:10px">Please come back in October! And now see [[http://nlp.ipipan.waw.pl/NLP-SEMINAR/previous-e.html|the talks given between 2000 and 2015]] and [[http://zil.ipipan.waw.pl/seminar|2015-16]]. | ||<style="border:0;padding-bottom:10px">'''NOTE''': Due to restriction of admission to the Institute building, only staff and speakers (including external ones) may currently take part in the seminar. For all other participants the seminar will be broadcast on [[https://www.youtube.com/ipipan|YouTube]].|| ||<style="border:0;padding-top:5px;padding-bottom:5px">'''5 October 2020'''|| ||<style="border:0;padding-left:30px;padding-bottom:0px">'''Piotr Rybak''', '''Robert Mroczkowski''', '''Janusz Tracz''' (ML Research at Allegro.pl), '''Ireneusz Gawlik''' (ML Research at Allegro.pl & AGH University of Science and Technology)|| ||<style="border:0;padding-left:30px;padding-bottom:5px">[[https://www.youtube.com/watch?v=LkR-i2Z1RwM|{{attachment:seminarium-archiwum/youtube.png}}]] '''[[attachment:seminarium-archiwum/2020-10-05.pdf|Review of BERT-based Models for Polish Language]]'''  {{attachment:seminarium-archiwum/icon-pl.gif|Delivered in Polish.}}|| ||<style="border:0;padding-left:30px;padding-bottom:15px">In recent years, a series of BERT-based models improved the performance of many natural language processing systems. During this talk, we will briefly introduce the BERT model as well as some of its variants. Next, we will focus on the available BERT-based models for Polish language and their results on the KLEJ benchmark. Finally, we will dive into the details of the new model developed in cooperation between ICS PAS and Allegro.|| ||<style="border:0;padding-top:5px;padding-bottom:5px">'''19 October 2020'''|| ||<style="border:0;padding-left:30px;padding-bottom:0px">'''Inez Okulska''' (NASK National Research Institute)|| ||<style="border:0;padding-left:30px;padding-bottom:5px">'''Concise, robust, sparse? Algebraic transformations of word2vec embeddings versus precision of classification'''  {{attachment:seminarium-archiwum/icon-pl.gif|Talk delivered in Polish.}}|| ||<style="border:0;padding-left:30px;padding-bottom:15px">The introduction of the vector representation of words, containing the weights of context and central words, calculated as a result of mapping giant corpora of a given language, and not encoding manually selected, linguistic features of words, proved to be a breakthrough for NLP research. After the first delight, there came revision and search for improvements - primarily in order to broaden the context, to handle homonyms, etc. Nevertheless, the classic embeddinga still apply to many tasks - for example, content classification - and in many cases their performance is still good enough. What do they code? Do they contain redundant elements? If transformed or reduced, will they maintain the information in a way that still preserves the original "meaning"? What is the meaning here? How far can these vectors be deformed and how does it relate to encryption methods? In my speech I will present a reflection on this subject, illustrated by the results of various "tortures” of the embeddings (word2vec and glove) and their precision in the task of classifying texts whose content must remain masked for human users.|| {{{#!wiki comment ||<style="border:0;padding-top:5px;padding-bottom:5px">'''2 April 2020'''|| ||<style="border:0;padding-left:30px;padding-bottom:0px">'''Stan Matwin''' (Dalhousie University)|| ||<style="border:0;padding-left:30px;padding-bottom:5px">'''Efficient training of word embeddings with a focus on negative examples'''  {{attachment:seminarium-archiwum/icon-pl.gif|Talk delivered in Polish.}} {{attachment:seminarium-archiwum/icon-en.gif|Slides in English.}}|| ||<style="border:0;padding-left:30px;padding-bottom:15px">This presentation is based on our [[https://pdfs.semanticscholar.org/1f50/db5786913b43f9668f997fc4c97d9cd18730.pdf|AAAI 2018]] and [[https://aaai.org/ojs/index.php/AAAI/article/view/4683|AAAI 2019]] papers on English word embeddings. In particular, we examine the notion of “negative examples”, the unobserved or insignificant word-context co-occurrences, in spectral methods. we provide a new formulation for the word embedding problem by proposing a new intuitive objective function that perfectly justifies the use of negative examples. With the goal of efficient learning of embeddings, we propose a kernel similarity measure for the latent space that can effectively calculate the similarities in high dimensions. Moreover, we propose an approximate alternative to our algorithm using a modified Vantage Point tree and reduce the computational complexity of the algorithm with respect to the number of words in the vocabulary. We have trained various word embedding algorithms on articles of Wikipedia with 2.3 billion tokens and show that our method outperforms the state-of-the-art in most word similarity tasks by a good margin. We will round up our discussion with some general thought s about the use of embeddings in modern NLP.|| }}} ||<style="border:0;padding-top:10px">Please see also [[http://nlp.ipipan.waw.pl/NLP-SEMINAR/previous-e.html|the talks given in 2000–2015]] and [[http://zil.ipipan.waw.pl/seminar-archive|2015–2020]].|| |
Natural Language Processing Seminar 2020–2021
The NLP Seminar is organised by the Linguistic Engineering Group at the Institute of Computer Science, Polish Academy of Sciences (ICS PAS). It takes place on (some) Mondays, normally at 10:15 am, in the seminar room of the ICS PAS (ul. Jana Kazimierza 5, Warszawa). All recorded talks are available on YouTube. |
|
NOTE: Due to restriction of admission to the Institute building, only staff and speakers (including external ones) may currently take part in the seminar. For all other participants the seminar will be broadcast on YouTube. |
5 October 2020 |
Piotr Rybak, Robert Mroczkowski, Janusz Tracz (ML Research at Allegro.pl), Ireneusz Gawlik (ML Research at Allegro.pl & AGH University of Science and Technology) |
In recent years, a series of BERT-based models improved the performance of many natural language processing systems. During this talk, we will briefly introduce the BERT model as well as some of its variants. Next, we will focus on the available BERT-based models for Polish language and their results on the KLEJ benchmark. Finally, we will dive into the details of the new model developed in cooperation between ICS PAS and Allegro. |
19 October 2020 |
Inez Okulska (NASK National Research Institute) |
Concise, robust, sparse? Algebraic transformations of word2vec embeddings versus precision of classification |
The introduction of the vector representation of words, containing the weights of context and central words, calculated as a result of mapping giant corpora of a given language, and not encoding manually selected, linguistic features of words, proved to be a breakthrough for NLP research. After the first delight, there came revision and search for improvements - primarily in order to broaden the context, to handle homonyms, etc. Nevertheless, the classic embeddinga still apply to many tasks - for example, content classification - and in many cases their performance is still good enough. What do they code? Do they contain redundant elements? If transformed or reduced, will they maintain the information in a way that still preserves the original "meaning"? What is the meaning here? How far can these vectors be deformed and how does it relate to encryption methods? In my speech I will present a reflection on this subject, illustrated by the results of various "tortures” of the embeddings (word2vec and glove) and their precision in the task of classifying texts whose content must remain masked for human users. |
Please see also the talks given in 2000–2015 and 2015–2020. |