+**********************************************************************
*
*
* Einladung
*
*
*
* Informatik-Oberseminar
*
*
*
+**********************************************************************
Zeit: Freitag, 12. Juli 2019, 10.00 Uhr
Ort: Informatikzentrum, E3, Raum 9222
Referent: Dipl.-Inform. Malte Nuhn
Thema: Unsupervised Training with Applications in Natural Language
Processing//
Abstract:
The state-of-the-art algorithms for various natural language processing
tasks require large amounts of labeled training data. At the same time,
obtaining labeled data of high quality is often the most costly step in
setting up natural language processing systems.Opposed to this,
unlabeled data is much cheaper to obtain and available in larger
amounts.Currently, only few training algorithms make use of unlabeled
data. In practice, training with only unlabeled data is not performed at
all. In this thesis, we study how unlabeled data can be used to train a
variety of models used in natural language processing. In particular, we
study models applicable to solving substitution ciphers, spelling
correction, and machine translation. This thesis lays the groundwork for
unsupervised training by presenting and analyzing the corresponding
models and unsupervised training problems in a consistent manner.We show
that the unsupervised training problem that occurs when breaking
one-to-one substitution ciphers is equivalent to the quadratic
assignment problem (QAP) if a bigram language model is incorporated and
therefore NP-hard. Based on this analysis, we present an effective
algorithm for unsupervised training for deterministic substitutions. In
the case of English one-to-one substitution ciphers, we show that our
novel algorithm achieves results close to human performance, as
presented in [Shannon 49].
Also, with this algorithm, we present, to the best of our knowledge, the
first automatic decipherment of the second part of the Beale
ciphers.Further, for the task of spelling correction, we work out the
details of the EM algorithm [Dempster & Laird + 77] and experimentally
show that the error rates achieved using purely unsupervised training
reach those of supervised training.For handling large vocabularies, we
introduce a novel model initialization as well as multiple training
procedures that significantly speed up training without hurting the
performance of the resulting models significantly.By incorporating an
alignment model, we further extend this model such that it can be
applied to the task of machine translation. We show that the true
lexical and alignment model parameters can be learned without any
labeled data: We experimentally show that the corresponding likelihood
function attains its maximum for the true model parameters if a
sufficient amount of unlabeled data is available. Further, for the
problem of spelling correction with symbol substitutions and local
swaps, we also show experimentally that the performance achieved with
purely unsupervised EM training reaches that of supervised training.
Finally, using the methods developed in this thesis, we present results
on an unsupervised training task for machine translation with a ten
times larger vocabulary than that of tasks investigated in previous work.
Es laden ein: die Dozentinnen und Dozenten der Informatik
_______________________________________________
--
--
Stephanie Jansen
Faculty of Mathematics, Computer Science and Natural Sciences
HLTPR - Human Language Technology and Pattern Recognition
RWTH Aachen University
Ahornstraße 55
D-52074 Aachen
Tel. Frau Jansen: +49 241 80-216 06
Tel. Frau Andersen: +49 241 80-216 01
Fax: +49 241 80-22219
sek(a)i6.informatik.rwth-aachen.de
www.hltpr.rwth-aachen.de
Tel: +49 241 80-216 01/06
Fax: +49 241 80-22219
sek(a)i6.informatik.rwth-aachen.de
www.hltpr.rwth-aachen.de
**********************************************************************
*
*
* Einladung
*
*
*
* Informatik-Oberseminar
*
*
*
+**********************************************************************
Zeit: Dienstag, 15. Dezember 2020, 10:00 Uhr
Zoom:
https://rwth.zoom.us/j/99233095930?pwd=dHhTV253V1ZYUzRtSkk1L3A1REZVUT09
Meeting-ID: 992 3309 5930
Kenncode: 626162
Referent: Philipp Weidel, Dipl. Inform.
Thema: Learning and decision making in closed loop simulations of
plastic spiking neural networks
Abstract:
To understand how animals and humans learn, form memories and make
decisions is a highly relevant goal both for neuroscience and for fields
that take some inspiration from neuroscience, such as machine learning
and artificial intelligence. Many models of learning and decision making
were developed in the fields of machine learning, artificial
intelligence, and computational neuroscience. Although these models aim
to describe similar mechanisms, they do not all pursue the same goal.
These models can be differentiated between models aiming to reach
optimal performance on a specific task (or set of tasks) and models
trying to explain how animals and humans learn. Some models of the first
class use biologically inspired methods (such as deep learning) but are
usually not biologically realistic and are therefore not well suited to
explain the function of the brain. Models in the second class focus on
being biologically plausible to explain how the brain works, but often
demonstrate their capability on too simplistic tasks and yield low
performance on well-known tasks from machine learning. This work aims to
close the gap between these two types of models.
In the first part of this talk, tools are described that allow the
combination of biologically plausible neural network models together
with powerful toolkits known from machine learning and robotics. To this
end, MUSIC, the middleware for spiking neural network simulators such as
NEST and NEURON is interfaced with ROS, a middleware for robotic
hardware and simulators such as Gazebo. This toolchain is extended with
interfaces to reinforcement learning toolkits such as the OpenAI Gym.
The second part addresses the question of how the brain can represent
its environment in the neural substrate of the cortex and how a
realistic model of reinforcement learning can make use of these
representations. To this end, a spiking neural network model of
unsupervised learning is presented which is able to learn its input
projections such that it can detect and represent repeating patterns. By
using an actor-critic reinforcement learning architecture driven by a
realistic dopamine modulated plasticity rule the model can make use of
the representations and learn a range tasks.
Es laden ein: die Dozentinnen und Dozenten der Informatik
--
Prof. Dr. Abigail Morrison
IAS-6 / INM-6 / SimLab Neuroscience
Jülich Research Center
&
Computer Science 3 - Software Engineering
RWTH Aachen
http://www.fz-juelich.de/inm/inm-6http://www.fz-juelich.de/ias/jsc/slnshttp://www.se-rwth.de
Office: +49 2461 61-9805
Fax # : +49 2461 61-9460
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDir Volker Rieke
Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
************************************************************************
*
*
* Einladung
*
*
*
* Informatik-Oberseminar
*
*
*
************************************************************************
Zeit: Montag, 21. Dezember 2020, 13:15-14:15 Uhr
Zoom: https://us02web.zoom.us/j/89686061674?pwd=UFUxaWMrdFJtY09JSWRBKzNxUnU0QT09
<https://www.google.com/url?q=https://us02web.zoom.us/j/89686061674?pwd%3DUF…>
Meeting-ID: 896 8606 1674
Kenncode: 923078
Referent: Alexander Hermans, M. Sc.
Lehrstuhl Informatik 13
Thema: Learning-based Visual Scene and Person Understanding for Mobile Robotics
Abstract:
We have seen tremendous progress in the computer vision community
across the past decades, especially with advancements in deep
learning, which is now used as the core machine learning approach for
most tasks. However, when deploying computer vision approaches in
actual robotic applications, we often find that top-performing methods
do not work as well as expected due to hardware constraints and
different input data characteristics.
We deal with visual scene and person understanding which are highly
relevant for robotics applications. Robots need to be able to understand
their environment and take special care around persons to ensure a safe
navigation and interaction. We specifically deal with three important
sub-tasks: semantic segmentation, 2D laser-based object detection, and
person re-identification. Semantic segmentation deals with the task of
labeling every pixel or point in a scene with a class label. This can in
turn be used to extract higher level information about the surrounding
scene, which can be used as context for further planning and interaction
tasks. While the resulting segmentations provide object labels, they do not
contain instance labels, making it hard to detect object instances.
However, object detection is an important capability for allowing robots to
safely navigate between dynamic objects. Especially the detection of
persons is an important task, enabling robots to interact with us. Since
many mobile platforms are already equipped with a 2D laser scanner, they
are interesting input sensors for object detection, even though the
resulting scans only contain sparse data. In addition to person detection,
person re-identification is an important task. This can be used to improve
tracking, but also allows to gather longer-term statistics and enables
person specific interactions.
For each of the three tasks we propose state-of-the-art approaches,
however, we also consider aspects that are important for their real-world
deployability and show applications of our methods within the context of
robotics projects.
Es laden ein: die Dozentinnen und Dozenten der Informatik
+**********************************************************************
*
*
* Einladung
*
*
*
* Informatik-Oberseminar
*
*
*
+**********************************************************************
Zeit: Donnerstag, 17. Dezember 2020, 15:00-16:00 Uhr
Zoom:
https://rwth.zoom.us/j/92609933435?pwd=OStsV2w0UUZrK2NmNnE4WDE0Y0pEQT09
Meeting-ID: 926 0993 3435
Kenncode: 584723
Referentin: Frau Can Sun, M. Sc.
Lehrstuhl Informatik 5
Thema: Embedding of real-world complexity modeling in adaptive supply chain
systems engineering
Abstract:
Successful supply chain complexity and change management is critical for
companies to gain competitive advantages. This thesis aims to build a
comprehensive supply chain complexity measurement framework through detailed
modeling and apply it in the changeable environment.
The research starts with modeling the supply chain and its dynamic change as
sociotechnical systems. Then a complexity measurement framework that is
decomposed into different levels of complexity drivers is developed, and a
decision-making framework for various change scenarios is proposed by
highlighting complexity as one of the decisive evaluation criteria.
The proposed solutions are validated with a set of real-world case studies
from the semiconductor industry. Each case represents one change scenario
and addresses the complexity from a particular perspective; the problem is
analyzed and modeled with the support of proposed models and frameworks. The
results show that complexity can be measured by a set of indicators or
formulas, and thus support decision-making by comparing the change-induced
complexity.
Es laden ein: die Dozentinnen und Dozenten der Informatik
_______________________________
Leany Maaßen
RWTH Aachen University
Lehrstuhl Informatik 5, LuFG Informatik 5
Prof. Dr. Stefan Decker, Prof. Dr. Matthias Jarke,
Prof. Gerhard Lakemeyer Ph.D.
Ahornstrasse 55
D-52074 Aachen
Tel: 0241-80-21509
Fax: 0241-80-22321
E-Mail: maassen(a)dbis.rwth-aachen.de
Dear all,
This is a short reminder about the talk of Kuldeep Meel (NUS Singapore)<https://www.unravel.rwth-aachen.de/cms/UnRAVeL/Das-Graduiertenkolleg/Gastwi…> today at 10:30.
Title: Sparse Hashing for Scalable Approximate Model Counting: When Theory and Practice Finally Meet
Given a Boolean formula F, the problem of model counting, also referred to as #SAT, is to compute the number of solutions of F. The hashing-based techniques for approximatecounting have emerged as a dominant approach, promising achievement of both scalability and rigorous theoretical guarantees. The standard construction of strongly 2-universal hash functions employs dense XORs (i.e., involving half of the variables in expectation), which is widely known to cause degradation in the runtime performance of state of the art SAT solvers.
Consequently, the past few years have witnessed an intense activity in the design of sparse XORs as hash functions. In this talk, we will first survey the known results, and identify the crucial bottleneck contributing to their lack of ability to provide speedups. We will then formalize a relaxation of universal hashing, called concentrated hashing, and establish a novel and beautiful connection between concentration measures of these hash functions and isoperimetric inequalities on boolean hypercubes. This allows us to obtain tight bounds on variance as well as the dispersion index and show that logarithmically sized XORs suffices for the design of sparse hash functions belonging concentrated hash family. Finally, we use sparse hash functions belonging to this concentrated hash family to develop new approximate counting algorithms. Our comprehensive experimental evaluation of our algorithm on 1896 benchmarks with computational effort of over 20,000 computational hours demonstrates speedup compared to existing approaches. To the best of our knowledge, this work is the first study to demonstrate runtime improvement of approximate model counting algorithms through the usage of sparse hash functions, while still retaining strong theoretical guarantees.
(Based on Joint work with S. Akshay, D. Agarwal, and Bhavishya; The corresponding papers were published at SAT-20 and LICS-20)
Bio:
Kuldeep Meel is Sung Kah Kay Assistant Professor of Computer Science in School of Computing at the National University of Singapore. He received his Ph.D. (2017) and M.S. (2014) degree from Rice University, and B. Tech. (with Honors) degree (2012) in Computer Science and Engineering from Indian Institute of Technology, Bombay. He is a recipient of 2019 NRF Fellowship for AI. His research interests lie at the intersection of Artificial Intelligence and Formal Methods. His work received the 2018 Ralph Budd Award for Best PhD Thesis in Engineering, 2014 Outstanding Masters Thesis Award from Vienna Center of Logic and Algorithms and Best Student Paper Award at CP 201
Wednesday, 2.12.2020, 10:30
https://rwth.zoom.us/j/92047949381?pwd=LzIwUW96WEM0MkRjZ01FUmhwd1I3QT09<https://www.google.com/url?q=https://rwth.zoom.us/j/92047949381?pwd%3DLzIwU…>
Meeting ID: 920 4794 9381
Password: unravel
https://www.unravel.rwth-aachen.de/go/id/kfsgc
Best regards
Helen Bolke-Hermanns
Helen Bolke-Hermanns
RTG UnRAVeL - RWTH Aachen University
Ahornstr. 55, D-52074 Aachen
Building E3, 2nd floor, Room 9218
Telefon: +49 (241) 80-21 004
Fax: +49 (241) 80-22 215
E-Mail: Helen.Bolke-Hermanns(a)Informatik.RWTH-Aachen.de<mailto:Helen.Bolke-Hermanns@Informatik.RWTH-Aachen.de>
Web: www.unravel.rwth-aachen.de<http://www.unravel.rwth-aachen.de/>
[rwth_unravel_en_rgb]