CMNS-UNDERGRAD-NEWS Archives

Daily eNews for CMNS Students

CMNS-UNDERGRAD-NEWS@LISTSERV.UMD.EDU

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Sender:
Daily eNews for CMNS Students <[log in to unmask]>
X-To:
Date:
Mon, 4 Mar 2019 11:28:51 -0500
Reply-To:
Daily eNews for CMNS Students <[log in to unmask]>
Subject:
MIME-Version:
1.0
Message-ID:
Content-Transfer-Encoding:
8Bit
Content-Type:
text/plain; charset=UTF-8; format=flowed; delsp=yes
From:
Sarah Kurtz <[log in to unmask]>
Parts/Attachments:
text/plain (46 lines)
Subject: Human Language Technology Center of Excellence (HLTCOE) - Paid
Internship

Description:
We are looking for outstanding undergraduate and graduate students for summer
internships in applied machine learning. The workshop is a good opportunity
for undergraduates to obtain research experience and for graduate students to
pursue challenging technical problems in a collaborative environment.
Previous workshops have resulted in fruitful collaborations beyond the
workshop itself and academic publications at top international conferences.

The theme of this year’s SCALE is automatically extracting structured
information (e.g. people, organizations, locations, etc.) from unstructured
text and, in particular, dealing with the challenges that emerge when labeled
training data is not readily available. This problem is known as named-entity
recognition and more broadly falls under the scope natural language
processing.

Some example technical problems in the scope of the workshop:

●       Fast and accurate structured prediction with deep neural models

●       Developing models that are robust to errors in the training data

●       Transferring language models trained on unlabeled text to target
domains

●       Using differentiable caches for “one-shot” learning from user
feedback

●       Semi-supervised training from partially labeled data

●       Multilingual training to alleviate sparsity in any one language

●       Identifying and exploiting incidental sources of training data

Experience with machine learning software packages, e.g. TensorFlow or
PyTorch, is a plus but not required.

APPLICATION DEADLINE: Monday, April 1, 2019
Acceptance decisions will be made on a rolling basis.

Contact Person: Raquel Robinson
Contact Email: [log in to unmask]
Website URL: https://hltcoe.jhu.edu/research/scale/

ATOM RSS1 RSS2