Moses-support Digest, Vol 125, Issue 29

Send Moses-support mailing list submissions to
moses-support@mit.edu

To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.mit.edu/mailman/listinfo/moses-support
or, via email, send a message with subject or body 'help' to
moses-support-request@mit.edu

You can reach the person managing the list at
moses-support-owner@mit.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Moses-support digest..."


Today's Topics:

1. Please help me to setup moses! (Ng? Th? Vinh)
2. [CFP] 2nd Call: WMT 2017 Shared Task on Bandit Learning for
Machine Translation (Artem Sokolov)


----------------------------------------------------------------------

Message: 1
Date: Wed, 15 Mar 2017 21:29:59 +0700
From: Ng? Th? Vinh <ntvinh@ictu.edu.vn>
Subject: [Moses-support] Please help me to setup moses!
To: moses-support@mit.edu
Message-ID:
<CA+VYTDirqbHh=E7V7r7kdxY-T61PRhK7+TcdJC7Y4R270_Fejw@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hello moses team,
I am a newer in moses, I compile SRILM on mac os el capitan but get an
error:

make[2]: [/Users/admin/translation/srilm/bin/macosx/maxalloc] Error 1
(ignored)

c++ -Wall -Wno-unused-variable -Wno-uninitialized -Wno-overloaded-virtual
-DINSTANTIATE_TEMPLATES -I/opt/local/libexec/macports/include -I.
-I/Users/admin/translation/srilm/include -DHAVE_ZOPEN -fPIC -u _matherr
-L/Users/admin/translation/srilm/lib/macosx -g -O2 -fno-common -o
../bin/macosx/ngram ../obj/macosx/ngram.o ../obj/macosx/liboolm.a
/Users/admin/translation/srilm/lib/macosx/libflm.a
/Users/admin/translation/srilm/lib/macosx/libdstruct.a
/Users/admin/translation/srilm/lib/macosx/libmisc.a
-L/opt/local/libexec/macports/lib/tcl8.5 -ltcl8.5 -lm -liconv 2>&1 | c++filt

Undefined symbols for architecture x86_64:

"File::offset(std::basic_ostream<char, std::char_traits<char> >&)",
referenced from:

NgramCounts<float>::readBinaryNode(Trie<unsigned int, float>&,
unsigned int, unsigned int, File&, long long&, bool, Array<unsigned int>&)
in liboolm.a(ClassNgram.o)

Ngram::readBinaryNode(Trie<unsigned int, BOnode>&, unsigned int,
unsigned int, File&, long long&, bool, Array<unsigned int>&) in
liboolm.a(NgramLM.o)

Ngram::skipToNextTrie(File&, unsigned int) in liboolm.a(NgramLM.o)

Ngram::readBinaryV1Node(Trie<unsigned int, BOnode>&, File&, File&,
bool, Array<unsigned int>&, unsigned int) in liboolm.a(NgramLM.o)

Ngram::writeBinaryNode(Trie<unsigned int, BOnode>&, unsigned int,
File&, long long&) in liboolm.a(NgramLM.o)

Ngram::writeBinaryV1Node(Trie<unsigned int, BOnode>&, File&, File&,
long long&, unsigned int) in liboolm.a(NgramLM.o)

Ngram::writeBinaryV1(File&) in liboolm.a(NgramLM.o)

...

"File::position(std::basic_ostream<char, std::char_traits<char> >&)",
referenced from:

RefList::read(File&, bool) in liboolm.a(RefList.o)

ClassNgram::readClasses(File&) in liboolm.a(ClassNgram.o)

NgramCounts<float>::read(File&, unsigned int, bool) [clone .part.83]
in liboolm.a(ClassNgram.o)

NgramCounts<float>::readBinary(File&, unsigned int, bool) in
liboolm.a(ClassNgram.o)

PQCodebook::read(File&) in liboolm.a(Prob.o)

AdaptiveMix::read(File&, bool) in liboolm.a(AdaptiveMix.o)

HMMofNgrams::read(File&, bool) in liboolm.a(HMMofNgrams.o)

I think it can not found some funntion:
NgramCounts<float>::readBinaryNode(), Ngram::skipToNextTrie(),.....

I do not know to get these funtions from what library?

Please tell me about error!
Thank you very much!

--
*Ng? Thi? Vinh*
Faculty of Electronics and Communications,
Thai Nguyen University of Information and Communication Technology (ICTU).
TEL: 0987 706 830
Email: *ntvinh@ictu.edu.vn <ptnghia@ictu.edu.vn>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20170315/167d8920/attachment.html

------------------------------

Message: 2
Date: Mon, 13 Mar 2017 17:44:58 +0100
From: Artem Sokolov <gposhta@gmail.com>
Subject: [Moses-support] [CFP] 2nd Call: WMT 2017 Shared Task on
Bandit Learning for Machine Translation
To: moses-support@mit.edu
Message-ID:
<CAAi4b7xHL2s4mhK-e36mdG-YpLSqEKuWWFVRMVradf_qUbYYmw@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Updates:

* Registration extended to March 28
* GitHub repository accessible to registered participants with SDK, code
examples and FAQ
* Mock service is now running! You can test your client's API calls
* Pre-processing and other details in a FAQ available on GitHub




CALL FOR PARTICIPATION
===============================================================
WMT 2017 Shared Task on Bandit Learning for Machine Translation
===============================================================
(collocated with EMNLP 2017)

Check the website for details: http://www.statmt.org/wmt17/ba
ndit-learning-task.html



###### BANDIT LEARNING FOR MACHINE TRANSLATION ######

Bandit Learning for MT is a framework to train and improve MT systems by
learning from weak or partial feedback: Instead of a gold-standard
human-generated translation, the learner only receives feedback to a single
proposed translation (this is why it is called partial), in form of a
translation quality judgement (which can be as weak as a binary
acceptance/rejection decision).

Amazon and University of Heidelberg organize this Shared Task with a goal to
encourage researchers to investigate algorithms for learning from weak user
feedback instead of from human references or post-edits that require skilled
translators. We are interested in finding systems that learn efficiently and
effectively from this type of feedback, i.e. they learn fast and achieve
high
translation quality. Developing such algorithms is interesting for
interactive
machine learning and for human feedback in NLP in general.

In the WMT task setup, the user feedback will be simulated by a service
hosted
on Amazon Web Services (AWS), where participants can submit translations and
receive feedback and use this feedback for training an MT model. Reference
translations will not be revealed at any point, also evaluations are done
via
the service.

###### IMPORTANT DATES ######

All dates are preliminary.

Registration via e-mail till March 28, 2017
Access to mock service March 13, 2017
Access to development service March 28, 2017
Online learning starts April 25, 2017
Notification of evaluation results May 26, 2017
Paper submission deadline June 9, 2017
Notification of paper acceptance June 30, 2017
Camera-ready deadline July 14, 2017

###### WHY IS IT CALLED BANDIT LEARNING? ######

The name bandit is inherited from a model where in each round a gambler in a
casino pulls an arm of a different slot machine, called "one-armed bandit",
with the goal of maximizing his reward relative to the maximal possible
reward,
without apriori knowledge of the optimal slot machine. In MT, pulling an arm
corresponds to proposing a translation; rewards correspond to user feedback
on
translation quality. Bandit learners can be seen as one-state Markov
Decision
Processes (MDPs), which connects them to reinforcement learning. In MT,
proposing a translation corresponds to choosing an action.

###### ONLINE LEARNING PROTOCOL ######

Bandit learning follows an online learning protocol, where on each of a
sequence of iterations, the learner receives a source sentence, predicts a
translation, and receives a reward in form of a task loss evaluation of the
predicted translation. The learner does not know what the correct prediction
looks like, nor what would have happened if it had predicted differently.

For t = 1, ..., T do
Receive source sentence
Predict translation
Receive feedback to predicted translation
Update system

Online interaction is done via accessing an AWS-hosted service that provides
source sentences to the learner (step 1), and provides feedback (step 3) to
the
translation predicted by the learner (step 2). The learner updates his
parameters using the feedback (step 4) and continues to the next example.

###### DATA ######

For training seed systems, out-of-domain parallel data shall be restricted
to
German-English Europarl, NewsCommentary, CommonCrawl and Rapid data for the
News Translation (constrained) task, monolingual English data from the
constrained task is allowed. Tuning of the out-of-domain system should be
done
on the newstest2016-deen development set. It is recommended to use Moses'
scripts for data pre-processing (normalization, tokenization, lowercasing).

The in-domain sequence of data for online learning will be e-commerce domain
provided by Amazon (pre-processed as above). Since the data comes from a
substantially different domain, expect a large number of out-of-vocabulary
terms. These data can only be accessed via the service. No reference
translations will be revealed, only feedback to submitted translations is
returned from the service.

Simulated reward-type real-valued feedback will be based on a combination of
several quality models, including automatic measures w.r.t. human
references,
and will be normalized to the range [0,1] ('very bad' to 'excellent').
Feedback
can only be accessed via the service. Only one feedback is allowed per
source
sentence.

###### SERVICES ######

Three AWS-hosted services will be provided:

* MOCK service to test client API: Will sample from a tiny in-domain
dataset and
simply return BLEU as feedback.
* DEVELOPMENT service to tune algorithms and hyperparameters: Will sample
from
a larger in-domain dataset. Feedback will be the BLEU measure. Several
runs
will be allowed and evaluation results will be communicated to the
participants.
* ONLINE Learning service: Will sample from a very large in-domain dataset.
Participants will have to consume a fixed number of samples during the
allocated online learning period to be eligible for final evaluation.
Feedback will be parameterized differently from the development service.

The respective data samples will be the same for all participants.

###### EVALUATION ######

The following main evaluation metrics will be used:
* ONLINE: cumulative per-sentence reward against the number of iterations,
* OFFLINE: standard automatic MT evaluation metric on a held-out in-domain
test set,
* RELATIVE to the out-of-domain starting point by doing test set
evaluations in the beginning and in the end of the online learning
sequence.

Note that all evaluations are done during online learning and not in a
separate
offline testing phase.

###### HOW TO PARTICIPATE ######

* Pick your favourite MT system.
* Train an out-of-domain model on allowed data.
* Register for the task via email and receive further instructions on how
to access the service.
* Wrap client code snippets around your MT system.
* Setup: Test the in-domain-training procedure with the mock service and
ensure that your client
sends translations and receives feedback.
* Tune: Find a clever strategy and good hyperparameters to learn from weak
feedback (e.g. by
simulating weak feedback from parallel data, or by using the development
service).
* Train your in-domain model by starting from your out-of-domain model,
submitting translations
to the online learning service, receiving feedback and updating your
model from this feedback.

###### ORGANIZERS ######

Pavel Danchenko, Amazon Development Center Berlin, Germany
Hagen Fuerstenau, Amazon Development Center Berlin, Germany
Julia Kreutzer, Heidelberg University, Germany
Stefan Riezler, Heidelberg University, Germany
Artem Sokolov, Heidelberg University and Amazon Development Center Berlin,
Germany
Kellen Sunderland, Amazon Development Center Berlin, Germany
Witold Szymaniak, Amazon Development Center Berlin, Germany
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20170313/dbac8efb/attachment.html

------------------------------

_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support


End of Moses-support Digest, Vol 125, Issue 29
**********************************************

Tweet