71. Theorietag - TU Dortmund...71. Theorietag Program Testingclusterstructureofgraphs Pan Peng (TU...

71. TheorietagFebruary 15 — February 16, 2016

Program and Workshop Information

71. Theorietag Program

Program Theorietag

Monday 15th, 2016

13:00 Registration (Foyer), Lunch

13:55-14:00

Opening

14:00-14:25

Amer KrivošijaClustering Time Series under the Fréchet distance

14:25-14:50

Pan PengTesting cluster structure of graphs

14:50-15:15

Gaetano GeckParallel-Correctness for Conjunctive Queries with Union and

Negation

15:15-15:45

Coffee Break

15:45-16:10

Daniel KönigThe circuit evaluation problem for finite semirings

16:10-16:35

Erik Jan van LeeuwenOn the Path to Independence

16:35-17:00

Stefan WalzerCard-based Cryptographic Protocols Using a Minimal Number of

Cards

19:00 Dinner at “Kumpel Erich”

ii


Tuesday 16th, 2016

9:00-10:00

Martin GroheThe Graph Isomorphism Problem

10:00-10:30

Coffee Break

10:30-10:55

Lisa ReyWinner Determination and Manipulation in Minisum and

Minimax Committee Elections

10:55-11:20

Anja ReyAltruistic Hedonic Games

11:20-11:45

Ann-Kathrin SelkerComplexity of Control in Judgment Aggregation for Uniform

Premise-Based Quota Rules

11:45-12:15

Coffee Break

12:15-12:40

Florian KurpiczOn the Benefit of Merging Suffix Array Intervals for Parallel

Pattern Matching

12:40-13:05

Dominik KöpplLempel Ziv Computation In Compressed Space (LZ-CICS)

afterwards Closing, Lunch, Departure

iii


Contents

Program 71. Theorietag ii

Overview of Talks 1

Workshop Information 9

Travel Information 11

Local Organization:Beate BolligJohannes FischerAnja FlehmigThomas SchwentickChristian SohlerNils VortmeierMatthias Westermann

Photo:Wolfgang Hunscher

iv


Overview of Talks

The Graph Isomorphism Problem

Martin Grohe (RWTH Aachen)

The question of whether there is a polynomial time algorithm deciding whether twographs are isomorphic has been a one of the best known open problems in theoreticalcomputer science for more than forty years. Indeed, the graph isomorphism problemis one of the very few natural problems in NP that is neither known to be in P norknown to be NP-complete. Very recently, Babai gave a quasipolynomial time isomor-phism algorithm. Despite of this breakthrough result, the question for a polynomialalgorithm remains wide open.My talk will be a survey of recent progress on the isomorphism problem. I will focuson generic algorithmic strategies (as opposed to algorithms tailored towards specificgraph classes) that have proved to be useful and interesting in various context, boththeoretical and practical.

Clustering time series under the Fréchet distance

Amer Krivošija (TU Dortmund)

Joint work of Anne Driemel, Amer Krivošija, Christian SohlerMain reference Anne Driemel, Amer Krivosija, Christian Sohler, “Clustering time

series under the Fréchet distance,” SODA 2016: 766–785URL http://dx.doi.org/10.1137/1.9781611974331.ch55URL http://arxiv.org/abs/1512.04349

The Fréchet distance is a popular distance measure for curves. We study the problemof clustering time series under the Fréchet distance. In particular, we give (1 + ε)-approximation algorithms for variations of the following problem with parameters kand `. Given n univariate time series P , each of complexity at most m, we find ktime series, not necessarily from P , which we call cluster centers and which eachhave complexity at most `, such that

(i) the maximum distance of an element of P to its nearest cluster center or

(ii) the sum of these distances

is minimized. Our algorithms have running time near-linear in the input size forconstant ε, k and `. To the best of our knowledge, our algorithms are the firstclustering algorithms for the Fréchet distance which achieve an approximation factorof (1 + ε) or better.

1


Testing cluster structure of graphs

Pan Peng (TU Dortmund)

Joint work of Artur Czumaj, Pan Peng, Christian SohlerMain reference Artur Czumaj, Pan Peng, and Christian Sohler, “Testing Cluster

Structure of Graphs,” In Proceedings of the Forty-Seventh AnnualACM on Symposium on Theory of Computing (STOC ’15), pages723–732.

URL http://dx.doi.org/10.1145/2746539.2746618

Graph clustering is a fundamental task in data analysis that aims to partition thevertex set of graph into a number of clusters, which are maximal subsets of verticesthat are well-connected to each other. In this talk, we will use the concept ofconductance to measure the quality of cluster structure and will focus on a questionof approximately recognizing cluster structure of a graph in sublinear time in theframework of property testing in the bounded degree model. We show how to testin O(

√n) time whether a graph with n nodes can be partitioned into no more than

k parts (clusters) such that the outer-conductance of each cluster is at most φo andthe inner-conductance of the induced subgraph on each cluster is at least φi, fora large spectrum of parameters k, φi, and φo. By the lower bound of Ω(

√n) for

testing graph expansion, which corresponds to the case when k = 1 in our problem,our algorithm is asymptotically optimal up to polylogarithmic factors.

Parallel-Correctness for Conjunctive Queries with Union and Negation

Gaetano Geck (TU Dortmund)

Joint work of Tom J. Ameloot, Gaetano Geck, Bas Ketsman, Frank Neven,Thomas Schwentick

Main reference Tom J. Ameloot, Gaetano Geck, Bas Ketsman, Frank Neven,Thomas Schwentick, “Parallel-Correctness and Transferability forConjunctive Queries,” PODS 2015: 47–58

URL http://doi.acm.org/10.1145/2745754.2745759Main reference Gaetano Geck, Bas Ketsman, Frank Neven, Thomas Schwentick,

“Parallel-Correctness and Containment for Conjunctive Querieswith Union and Negation,” accepted at ICDT 2016

URL http://arxiv.org/abs/1412.4030

Single-round multiway join algorithms first reshuffle data over many servers and thenevaluate the query at hand in a parallel and communication-free way. A key question iswhether a given distribution policy for the reshuffle is adequate for computing a givenquery, also referred to as parallel-correctness. We study the complexity of parallel-correctness for conjunctive queries and extend this study to unions of conjunctivequeries with and without negation.

2


The circuit evaluation problem for finite semirings

Daniel König (Universität Siegen)

Joint work of Moses Ganardi, Danny Hucke, Daniel König, Markus Lohrey

The computational complexity of the circuit evaluation problem for finite semiringsis considered, where semirings are not assumed to have an additive or multiplicativeidentity. The following dichotomy is shown: If a finite semiring is such that

(i) the multiplicative semigroup is solvable and

(ii) it does not contain a subsemiring with an additive identity 0 and a multiplicativeidentity 1 6= 0,

then the circuit evaluation problem for the semiring is in DET ⊆ NC2. In all othercases, the circuit evaluation problem is P-complete.

On the Path to Independence

Erik Jan van Leeuwen (Max-Planck Institut für Informatik, Saarbrücken)

Joint work of Daniel Lokshtanov, Marcin Pilipczuk, Erik Jan van LeeuwenMain reference Daniel Lokshtanov, Marcin Pilipczuk, Erik Jan van Leeuwen,

“Independence and Efficient Domination on P6-free Graphs,” inProceedings of the Twenty-Seventh Annual ACM-SIAMSymposium on Discrete Algorithms: 1784–1803

URL http://dx.doi.org/10.1137/1.9781611974331.ch124URL http://arxiv.org/abs/1507.02163

The Independent Set problem is a fundamental and well-studied graph problem, andasks to find a largest set of pairwise nonadjacent vertices of a graph. AlthoughIndependent Set is known to be NP-hard or worse for many graph classes, surprisingly,the complexity of Independent Set on graphs that exclude a fixed graph H as aninduced subgraph (so-called H-free graphs) is not yet settled. While Independent Setis still NP-hard for H-free graphs for most fixed graphs H, the complexity is openwhen H is a path of 6 or more vertices (it is polynomial when H is a shorter path).In this talk, I will survey the known results about Independent Set on H-free graphs.In particular, I will speak about recent work with Daniel Lokshtanov and MarcinPilipczuk, where we show that when H is a path on 6 vertices, Independent Set hasa quasipolynomial-time algorithm. Hence, in this case, a simultaneous NP-hardnessresult would imply quasipolynomial-time algorithms for all problems in NP.

3


Card-based Cryptographic Protocols Using a Minimal Number of Cards

Stefan Walzer (TU Ilmenau)

Joint work of Alexander Koch, Stefan Walzer, Kevin HärtelMain reference Alexander Koch, Stefan Walzer, Kevin Härtel, “Card-Based

Cryptographic Protocols Using a Minimal Number of Cards,” InAdvances in Cryptology — ASIACRYPT (1) 2015: 783-807.

URL http://dx.doi.org/10.1007/978-3-662-48797-6_32

A playing card has either ♥ or ♣ on its front. All cards are blank on the back. It isnot surprising that information can be represented and computations can be carriedout if sequences of cards are interpreted and rearranged according to specific rules.What is more surprising is, that computations can be made securely, meaning thatwe (who carry out the computation) remain oblivious of both the input and output.When computing a Boolean AND, two particular protocols come to mind: Firstly, denBoer (EUROCRYPT ’89) devised his famous “five-card trick”, which hides the input,but not the output. Secondly, a six-card protocol by Mizuki and Sone (FAW 2009),which hides both. In their paper, the authors ask whether six cards are minimal.We give a comprehensive answer to this question: there is a four-card AND protocolwith a runtime that is finite in expectation (i.e. a Las Vegas protocol), but noprotocol with finite runtime. Moreover, we show that five cards are sufficient forfinite runtime.However, these results must be taken with a grain of salt: One can argue that theunderlying computational model does not accurately capture the non-deterministic“shuffle” operations that real people can do with real cards. We shall therefore brieflydiscuss an alternative as well.

Winner Determination and Manipulation in Minisum and MinimaxCommittee Elections

Lisa Rey (Heinrich-Heine-Universität Düsseldorf)

Joint work of Dorothea Baumeister, Sophie Dennisen, Lisa ReyMain reference Dorothea Baumeister, Sophie Dennisen, Lisa Rey, “Winner

Determination and Manipulation in Minisum and MinimaxCommittee Elections,” in Proceedings of the 4th InternationalConference on Algorithmic Decision Theory (ADT 2015), pages469–485


In a committee election, a set of candidates has to be determined as winner ofthe election. Baumeister and Dennisen [1] proposed to extend the minisum andminimax approach, initially defined for approval votes [2], to other forms of votes.They define minisum and minimax committee election rules for trichotomous votes,incomplete linear orders and complete linear orders, by choosing a winning committeethat minimizes the dissatisfaction of the voters. Minisum election rules minimizethe voter dissatisfaction by choosing a winning committee with minimum sum of the

4


disagreement values for all individual votes, whereas in a minimax winning committeethe maximum disagreement value for an individual vote is minimized. Here, weinvestigate the computational complexity of winner determination in these votingrules. We show that winner determination is possible in polynomial time for allminisum rules we consider, whereas it is NP-complete for three of the minimax rules.Furthermore, we study different forms of manipulation for these committee electionrules.

References

[1] D. Baumeister and S. Dennisen. Voter Dissatisfaction in Committee Elections. InProceedings of the 14th International Joint Conference on Autonomous Agentsand Multiagent Systems. IFAAMAS, 2015. Extended abstract.

[2] S. Brams, D. Kilgour, and R. Sanver. A minimax procedure for negotiatingmultilateral treaties. In M. Wiberg, editor, Reasoned choices: Essays in Honorof Hannu Nurmi. Finnish Political Science Association, 2004.

Altruistic Hedonic Games

Anja Rey (Heinrich-Heine-Universität Düsseldorf)

Joint work of Nhan-Tam Nguyen, Anja Rey, Lisa Rey, Jörg Rothe, Lena SchendMain reference Nhan-Tam Nguyen, Anja Rey, Lisa Rey, Jörg Rothe, Lena Schend,

“Altruistic Hedonic Games,” to appear in Proceedings of the 15thInternational Conference on Autonomous Agents and MultiagentSystems (AAMAS 2016)

Hedonic games, proposed by Drèze and Greenberg [4] and later formally modelled byBogomolnaia and Jackson [2] and Banerjee et al. [1], are coalition formation gamesin which players have preferences over the coalitions they can join. All models ofrepresenting hedonic games studied so far are based upon selfish players only. Amongthe known ways of representing hedonic games compactly, we focus on friend-orientedhedonic games [3] and propose a novel model for them that takes into accountnot only a player’s own preferences but also her friends’ preferences under threedegrees of altruism. We study both the axiomatic properties of these games and thecomputational complexity of problems related to various stability concepts such asthe existence and verification of Nash stability, core stability, and popularity.

References

[1] S. Banerjee, H. Konishi, and T. Sönmez. Core in a simple coalition formationgame. Social Choice and Welfare, 18(1):135–153, 2001.

5


[2] A. Bogomolnaia and M. Jackson. The stability of hedonic coalition structures.Games and Economic Behavior, 38(2):201–230, 2002.

[3] D. Dimitrov, P. Borm, R. Hendrickx, and S. Sung. Simple priorities and corestability in hedonic games. Social Choice and Welfare, 26(2):421–433, 2006.

[4] J. Drèze and J. Greenberg. Hedonic coalitions: Optimality and stability. Econo-metrica, 48(4):987–1003, 1980.

Complexity of Control in Judgment Aggregation for UniformPremise-Based Quota Rules

Ann-Kathrin Selker (Heinrich-Heine-Universität Düsseldorf)

Joint work of Dorothea Baumeister, Jörg Rothe, Ann-Kathrin SelkerMain reference Dorothea Baumeister, Jörg Rothe, Ann-Kathrin Selker,

“Complexity of Bribery and Control for Uniform Premise-BasedQuota Rules Under Various Preference Types,” in Proceedings ofthe 4th International Conference on Algorithmic Decision Theory(ADT 2015), pages 432–448


Judgment aggregation is a framework for collective decision making, where a num-ber of agents (called judges) aggregate their judgments on possibly interconnectedpropositions in order to determine a collective outcome. Therefore it is important tostudy the susceptibility of judgment aggregation procedures to (external or internal)influences. Even though procedures are often susceptible to certain kinds of influ-ence, computational complexity can act as a shield to hinder undesirable strategicbehaviour.This presentation focuses on control scenarios where an external agent tries to obtaina preferred collective outcome by changing the structure of the judgment process,e.g. by adding or deleting judges. Thus we study different preference types of theexternal agent, including closeness-respecting and Hamming-distance-induced pref-erences introduced by Dietrich and List (2007), and obtain NP-hardness results forseveral control problems regarding the family of uniform premise-based quota rules.

On the Benefit of Merging Suffix Array Intervals for Parallel PatternMatching

Florian Kurpicz (TU Dortmund)

Joint work of Johannes Fischer, Dominik Köppl, Florian Kurpicz

We present parallel algorithms for exact and approximate pattern matching with suffixarrays, using a CREW-PRAM with p processors. Given a static text of length n, wefirst show how to compute the suffix array interval of a given pattern of length min O(m

p+ lg lg p lg lgn) time for p ≤ m. For approximate pattern matching with

6


k differences or mismatches, we show how to compute all occurrences of a givenpattern in O(mk

pσk max (k, lg lgn) + occ) time, where σ is the size of the alphabet

and p ≤ σkmk. The workhorse of our algorithms is a data structure for merging suffixarray intervals quickly: Given the suffix array intervals for two patterns P and P ′, wepresent a data structure for computing the interval of PP ′ in O(lg lgn) sequentialtime, or in O(lgp lgn) parallel time. All our data structures are of size O(n) bits (inaddition to the suffix array).

Lempel Ziv Computation In Compressed Space (LZ-CICS)

Dominik Köppl (TU Dortmund)

Joint work of Dominik Köppl, Kunihiko SadakaneMain reference Dominik Köppl, Kunihiko Sadakane, “Lempel-Ziv Computation In

Compressed Space (LZ-CICS),” Accepted at Data CompressionConference 2016

URL http://arxiv.org/abs/1510.02882

We show that both the Lempel-Ziv-77 and the Lempel-Ziv-78 factorization of a textof length n on an integer alphabet of size σ can be computed in O(n lg lg σ) time(linear time if we allow randomization) using O(n lg σ) bits of working space.

7


Campus Map

Campus Nord1a. Maschinenbau (Pav. 10:EF 73)1b. Halle Fluidenergiemaschinen

(EF71b)1c. Referat Arbeits-, Umwelt- und

Gesundheitsschutz (EF 71a)2. Leitwarte, Blockheizkraftwerk

(EF 71c)3. Dez. 6 - THB (EF 71)4. Dez.4:Studierendenservice,ReferatInternationales,Dez. 1 (hsp), zhb(EF61)

4a. InternationalesBegegnungszentrum (IBZ) (EF 59)

5. Maschinenbauhalle (LE 1)6. Bio- und Chemieingenieurwesen,Maschinenbau, Elektrotechnik,Dez. 3, Stabstelle Chancengleich-heit, Familie und Vielfalt, Gleich-stellungsbüro, Schwerbehinderten-vertretung (EF 68/70)

6a. Wissenschaftl. Personalrat,Nicht-wissenschaftl. Personalrat, JAV,Dez. 6.1 (EF 72)

7. Studierendenwerk,Mensa (VP 85)8. Erziehungswissenschaft,Psy-chologie und Soziologie,Human-wissenschaften undTheologie,Rehabilitationswissenschaften,Kulturwissenschaften,Kunst- undSportwissenschaften, ITMC,AStA,DoKoLL, zhb.dobus (EF 50)

9. Unicenter, LehrredaktionJournalistik (VP 74)

10. Physik - DELTA (MGM 2)11a. Maschinenbau I (LE 5)11b. Maschinenbau II (LE 2)12. Chemie,WiSo, Elektrotechnik,

Maschinenbau,ZentraleVervielfältigung (OH 6)

13. Hörsaalgebäude II (OH 4)14. Audimax,Mathematik, Statistik,

Wirtschafts- und Sozialwissen-schaften (VP 87)

15. Universitätsbibliothek (VP 76)16. Statistik, Zentrum für

HochschulBildung (zhb), Institut fürSchulentwicklungsforschung (IFS)(CDI-Gebäude:VP 78)

17a. Informatik (OH 16)17b. Informatik (OH 14)17c. ITMC, Informatik (OH 12)18. Elektrotechnik und

Informationstechnik (FWW 4)19. Elektrotechnik, Institut für

Roboterforschung (OH 8)20. Wirtschafts- und Sozialwissen-

schaften (Pav. 11:OH 6a)21a. Physik, Elektrotechnik und

Informationstechnik,WiSo (OH 4)21b. Ersatzneubau Chemie-Physik

(im Bau) (OH 4a)22. Erich-Brost-Institut (OH 2)23. Campus Treff (VP 120)

24. Kunst- und Sportwissenschaften,Fitnessförderwerk (OH 3)

25. Seminarraumgebäude (FWW 6)26. Kindertagesstätte HoKiDo (EF 57)27. LogistikCampus (JF 2-4)28. A1–A3 Dez. 5 (MSW 12, 13, 16),

WiSo (MSW 12)

Campus Süd29. (GB V:AS 12)30. Raumplanung (GB III: AS 10)31. Architektur und Bauingenieurwesen

(GB II: AS 8)32. Raumplanung, Architektur und

Bauingenieurwesen (GB I:AS 6)33. Hörsäle, Rektorat, Kanzler, Dez. 1,

Referat Innenrevision (HG I:AS 4)33a. Modellbauwerkstatt (AS 4a)34. Dez. 2, Dez. 5, Referat Controlling

(WD 2)35. Dez. 3 (AS 1)36. (GB IV:BS 301)36a. Maschinenbau III (BS 303)37. Experimentierhalle (BS 299)38. Archeteria (AS 2)39a. Referat Hochschulkommunikation,

Referat Hochschulmarketing(BS 285)

39b. Referat Forschungsförderung undWissenstransfer (BS 283)

40. (Pav. 5:BS 281)41. Rudolf-Chaudoire-Pavillon (BS 297)

42. Lagerhalle (BS 299)43. Dezernat 3 (Pav. 8:WD 1)44. (Pav. 2a:WD 2a)45. Haus Dörstelmann, AStA

(Pav. 1:BS 322)46. (Pav. 7:BS 322)47. Helmut Keunecke Haus /

Gästehaus (BS 233)48. (Pav. 3:WD 6)49. (Pav. 2b:WD 4)50. (Pav. 4:BS 279)

LegendeAS August-Schmidt-StraßeBS Baroper StraßeEF Emil-Figge-StraßeFWW Friedrich-Wöhler-WegJF Joseph-von-Fraunhofer-StraßeLE Leonhard-Euler-StraßeMGM Maria-Goeppert-Mayer-StraßeMSWMartin-Schmeißer WegOH Otto-Hahn-StraßeVP VogelpothswegWD Wilhelm-Dilthey-StraßeP ParkplätzeH Haltestelle H-BahnH Haltestelle Bus und Bahn

A1-A3Anmietungen

P

P

PP

P

P

P

P

PP

P

P

P

P

P

P

36

28/A3

3

4 4a

5 67

8

10 11a

9

18

12

1920

21a

13

14

15 16

23

24

22

29

3333

32

34

38

40

37

35

2

3031

1a1b

1c

41

17b

45

46

47

39a39b

50

4844

49

26

611

43

26

42

25

2627

36a

17c

21b28/A2

28/A1

ab

17a

a

ISAS

nrwision

Freige-lände MB

Fraunhofer-Institut für Software- und Systemtechnik ISST

Fraunhofer-Institut für Materialfluss und Logistik

Max-Planck-Institut fürmolekulare Physiologie

Technologie-zentrum

F&E Gesellschaft

Voge

lpot

hsw

eg (V

P)

Baroper Stra

ße (BS)

August-Schmidt-Straße (AS)

Wilhelm-Dilthey-Stra

ße (WD)

Leon

hard

-Eul

er-S

traß

e (L

E)

Mar

tin-

Sch

mei

ßer W

eg (M

SW

)

Hau

ert

Ostenbergstraße

Am G

arde

nkam

p

Hugo-Heimsath-Str.

Jose

ph-v

on-F

raun

hofe

r Str

aße

B1

Maria-Goeppert-Mayer-Straße (MGM)

Otto-Hahn-Straße (OH)

Universitätsstraße

Am Paß

Frie

dric

h-W

öhle

r-W

eg

Martin-Schmeißer-

Platz

Vogelpothsweg

Bar

ope r

Str

a ße

Emil-Figge-Straße (EF)

H

H

H

H

H

S

HH

H

H

H H

H

H

Meitnerweg

Universität

Emil-Figge-Str.

Technologiez.

H-BahnCampus Süd

Eichlinghofen Am Gardenkamp

H-BahnEichlinghofen

Groß Barop

Technologie-zentrum

Joseph-von-FraunhoferStraße

H Meitnerweg/Wohnanlage

H

Martin-Schmeißer-Weg

H

Am Hedreisch

Außensportanlagen

CampusNord

CampusSüd

A B

C

D

E

E

EEE

A conference siteB bus station “Meitnerweg”C train station S1

D parking,E refectories,

restaurants

8


Workshop Information

VenueThe workshop site is located in the computer science building at the Otto-Hahn-Straße on the North Campus (Campus Nord) of the Technische Universität (TU)Dortmund.

Address: Otto-Hahn-Straße 12, 44227 Dortmund

The workshop will take place in room E.003 which is located on the ground floor ofthe building.

Registration deskThe registration desk is in the foyer of the computer science building.phone: +49 (0) 231 755 6223 e-mail: [email protected]

Internet accessIf your university participates in “eduroam”, you have access to the correspondingwireless network. If you can not use this network, please ask for a guest account atthe registration desk.

LunchSome refectories and restaurants near the workshop site are marked on the CampusMap. The menus are available at the registration desk.

DinnerOn monday at 19:00, we will have dinner at “Kumpel Erich”, a restaurant in the“Kreuzviertel”, one of the most popular quarters of Dortmund. From the workshopsite, please follow the directions on page 11. The location is marked on the CityMap.Address: Kreuzstraße 87, 44137 Dortmund

9


City Maps

A

B

C

D

E

A subway station “Kreuzstraße”B restaurant “Kumpel Erich”C Hotel Steigenberger

Data from OpenStreetMap - published under ODbL

D subway station “Stadtgarten”E JGH Adolph Kolping

10


Travel Information

The workshop is held at the lecture hall E.003 at Otto-Hahn-Straße 12 on the northcampus (Campus Nord) of the university.

Arrival by trainFrom Dortmund main station take the train S1 in the direction of Düsseldorf to getto the station “DO-Universität” (point C on the Campus Map). (If your train stopsin Bochum, you can take the S1 from Bochum in the direction of Dortmund to get tothe station “DO-Universität”.) The workshop site is at a convenient walking distancefrom the train station (10 minutes).

Arrival by carParking lots are marked on the Campus Map.

From the university to the hotelsFrom the workshop site, take the bus number 445 or 462 from the station “Meitner-weg” (point B on the Campus Map) in direction “An der Palmweide”. At “An derPalmweide”, change to subway U42 towards “Grevel”. To get to Steigenberger Hotelor “Kumpel Erich”, exit at station “Kreuzstraße”. To get to JGH Adolph Kolping, exitat station “Stadtgarten”. All locations are marked on the City Maps (page 10).

Time table informationWe recommend this website for public transport time table information:http://efa.vrr.de, mobile version http://mobil.vrr.de

11

71. Theorietag - TU Dortmund...71. Theorietag Program Testingclusterstructureofgraphs Pan Peng (TU...

Documents

Transcript of 71. Theorietag - TU Dortmund...71. Theorietag Program Testingclusterstructureofgraphs Pan Peng (TU...