Julian Blank

Michigan State University, East Lansing, MI 48824, USA ·

Julian Blank received the B.Sc. in business information systems from Otto von Guericke University, Germany in 2010. He was a visiting scholar for six month at the Michigan State University, Michigan, USA in 2015, and, finished his M.Sc. in Computer Science at Otto von Guericke University, Germany in 2016. He is currently a Ph.D. student in Computer Science at the Michigan State University, Michigan, USA. His current research interests include multi-objective optimization, evolutionary computation, surrogate-assisted optimization, machine learning and deep learning.


Blank J., Deb K. and Roy P. (2019). Investigating the Normalization Procedure of NSGA-III. International Conference on Evolutionary Multi-Criterion Optimization (EMO) 2019. [link] [slides] [pdf] [source code] [bibtex]

Roy P, Hussein R., Blank J. and Deb K. (2019). Trust-Region Based Multi-objective Optimization for Low Budget Scenarios. International Conference on Evolutionary Multi-Criterion Optimization (EMO) 2019. [link] [slides] [pdf] [bibtex]

Vesikar Y., Deb K. and Blank J. (2018). Reference Point Based NSGA-III for Preferred Solutions. IEEE Symposium Series on Computational Intelligence (SSCI) 2018. [link] [pdf] [source code] [bibtex]

Talukder K., Deb K. and Blank J. (2018). Visualization of the boundary solutions of high dimensional pareto front from a decision maker's perspective. Proceedings of the Genetic and Evolutionary Computation Conference Companion. GECCO 2018. [link] [poster] [pdf] [bibtex]

Roy P. and Blank J. and Hussein R. and Deb K. (2018). Trust-region Based Algorithms with Low-budget for Multi-objective Optimization. Proceedings of the Genetic and Evolutionary Computation Conference Companion. GECCO 2018. [link] [pdf] [bibtex]

Blank J., Deb K., Mostaghim S. (2017). Solving the Bi-objective Traveling Thief Problem with Multi-objective Evolutionary Algorithms. Evolutionary Multi-Criterion Optimization. EMO 2017. Lecture Notes in Computer Science, vol 10173. Springer, Cham. [pdf] [bibtex]

Blank, J., Mostaghim S. and Deb K. (2016). In-Depth Analysis and Characteristics of the Traveling Thief Problem. Master's thesis, University of Magdeburg. [pdf] [bibtex]

Blank, J., Sander, F. (2015). Computational Intelligence in Games: The General Video Game AI Competition. Report, University of Magdeburg. [pdf]

Blank, J., Kruse R., Moewes C. and Buechner D. (2014). Evaluierung regelbasierter Klassifikatoren fuer eine Anwendung zur Fehlerregelgenerierung aus großen Datenmengen. Bachelor's thesis, University of Magdeburg. [pdf] [bibtex]



Multi-objective Test Problems

This framework provides a collection of test problems in Python. The main features are: Most important multi-objective test function is one place Vectorized evaluation by using numpy matrices Easily new problems can be created using custom classes or functions Here, you can find a detailed documentation and information about the framework.


Multi-objective Optimization Framework

Implementation of various multi-objective algorithms, such as NSGA-II, NSGA-III, MOEAD\D and so on.


Multi-Objective Optimization

Multi-Objective Optimization optimizes a vector of variable instead of only one. For this reason there are indifferent solutions.

Traveling Thief Problem

People are facing real-world problems with dependencies and interwovenness every day. In order to provide studies on an interwoven problem to fill that gap between theory and practice, the Traveling Thief Problem, where the Traveling Salesman Problem and Knapsack Problem interact, was proposed. Research done on this problem will bring insights how to handle problems with interwovenness and therefore how to solve problems with real-world character.


The thief has to effects when these two problems are combined:

  • When an item is picked the velocity decreases and the thief slows down.
  • There is a depreciation which increases the value of items over time.

Both effects are affecting both components and therefore it is a complete interwoven system. There are different versions of the problem which were propsed. Single-objective by weights, multi-objective with depreciation and single-objective by a minimal profit contstraint.

We created a benchmark in order to investigate the Traveling Thief Problem systematically. Here


Sudoku is a logic based game based on a 9x9 grid. All cells have to be filled with values between 1 and 9 whereby:

  • Each row contains each value only once
  • Each column contains each value only once
  • Each box contains each value only once

Many algorithms are proposed for solving this problem. One of the most efficient and best once is very close to the strategy of humans. My Java implements is provided at Github

Smart Groups

Smart Groups was a tool which I developed originally for a teacher. His school had every year the same problem in order to create groups out from the whole class. The task looked like the following:

  • Create subgroups out of the class. In this case the class had 17 pupils and he wanted to have three groups of 4 and one group of 5.
  • The teachers are allowed to define subsets which are not allowed to be in a group. For example if John is always fighting with Dan the teachers are allowed to set forbidden groups beforehand. Additionally, they are allowed to define the opposite which means to assign a subset which must be fulfilled in the final solution. In that case a disabled person should have been in the same group like the other one.
  • Each member is also allowed to give preferences to be with in a group. But also to define rejections to other members.

The solution should consider the constraints of the teacher as more important than the wishes of the members. But every member should be as satisfied as possible.


Work Experience

Teaching Assistant for Computer Organization and Architecture

Michigan State University

Teaching Assistant for Computer Organization and Architecture taught by Professor Owen. This class is about the design of combinational and sequential circuits and architecture and organization of digital computing systems. As a teaching assistants we were responsible for help rooms, answer questions online and grading the assignments as well as the midterm and final exams.

01/2019 - 05/2019

Collaborative Research for Structural Optimization

The project lasts for more than two years and included different optimization problems and runs. Two different practical simulation-based optimization problems were investigate and a surrogate-assisted method as well as an meta-model based optimization framework was proposed.

06/2017 - 12/2018

Java Enterprise Edition Developer

Full Stack Developer for innovative large-scale Business Applications using Java Enterprise Edition Technologies and Frameworks: Java EE (JPA, EJB, JSF, Primefaces, Omnifaces), MySQL, Jenkins, Jira

12/2016 - 05/2017

Research Assistant for Multi-objective Optimization

Michigan State University

Research Assistant for Multi-objective Optimization at the Computational Optimization and Innovation (COIN) Laboratory supervised by Professor Kalyanmoy Deb

08/2015 - 01/2016

Teaching Assistant for Data Warehouse Technologies

Otto von Guericke University Magdeburg

Teaching assistant for Data Warehouse Architecture, OLAP and practical exercises on Oracle Databases. The lecture was hold by Dr. Veit Köppen and Andreas Meister.

10/2014 - 01/2015

Software Developer

Software developer for a tool used in financial consulting. The project was managed by Scrum and the tool provided an automated data analyses for regression tests.

11/2013 - 03/2014

Research Assistant for SAP Education

Developing and maintaining educational material in the field of SAP. Implementing a mobile application in order to communicate with SAP ERP Systems.

06/2011 - 06/2013


Michigan State University

Computer Science
08/2017 - Today

Otto von Guericke University Magdeburg

Master of Science
Computer Science
04/2014 - 04/2016

Otto von Guericke University Magdeburg

Bachelor of Science
Business Information Systems
10/2010 - 05/2014