Difan Deng, M. Sc.

Phone

+49 511 762 5322

Fax

+49 511 762 5333

d.dengai.uni-hannover.de

Address

Appelstraße 9a
30167 Hannover

Building

3408

Room

1332

Phone

+49 511 762 5322

Fax

+49 511 762 5333

d.dengai.uni-hannover.de

Address

Appelstraße 9a
30167 Hannover

Building

3408

Room

1332

I am working towards a Ph.D. at Leibniz University Hannover. Previously I obtained my Master's degree in electrical engineering and information technology at TU Darmstadt and my bachelor's degree in electronics and information engineering at Huazhong University of Science in 2019 and 2015 respectively.

My research interest is AutoML, including Hyperparameter Optimization and Neural Architecture Search. The goal is to provide easy-to-use AutoML systems that allow non-expert Machine learning users to work with machine learning problems at hand.

Research Interests

Hyperparameter Optimization
Neural Architecture
Time Series forecasting

Curriculum Vitae

Education

Since 2020, PhD Candidate, University Hannover, Germany

2016-2019, Master of Science, Electrical Engineering and Information Technology , TU Darmstadt, Germany

2011-2015, Bachelor of Engineering, Electronic and Information Engineering, Huazhong University of Science and Technology

Publications

2023

Bischl, B., Binder, M., Lang, M., Pielok, T., Richter, J., Coors, S., Thomas, J., Ullmann, T., Becker, M., Boulesteix, A-L., Deng, D., & Lindauer, M. (2023). Hyperparameter Optimization: Foundations, Algorithms, Best Practices and Open Challenges. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery.

doi.org/10.1002/widm.1484

2022

Deng, D., Karl, F., Hutter, F., Bischl, B., & Lindauer, M. (2022). Efficient Automated Deep Learning for Time Series Forecasting. In Proceedings of the European Conference on Machine Learning (ECML)

doi.org/10.48550/arXiv.2205.05511

Deng, D., & Lindauer, M. (2022). Searching in the Forest for Local Bayesian Optimization. In ECML/PKDD workshop on Meta-learning

arxiv.org/abs/2111.05834

Lindauer, M., Eggensperger, K., Feurer, M., Biedenkapp, A., Deng, D., Benjamins, C., Sass, R., & Hutter, F. (2022). SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimization. Journal of Machine Learning Research.

arxiv.org/abs/2109.09831

Ruhkopf, T., Mohan, A., Deng, D., Tornede, A., Hutter, F., & Lindauer, M. (2022). MASIF: Meta-learned Algorithm Selection using Implicit Fidelity Information.

2021

Guerrero-Viu, J., Hauns, S., Izquierdo, S., Miotto, G., Schrodi, S., Biedenkapp, A., Elsken, T., Deng, D., Lindauer, M., & Hutter, F. (2021). Bag of Baselines for Multi-objective Joint Neural Architecture Search and Hyperparameter Optimization. In ICML 2021 Workshop AutoML

arxiv.org/abs/2105.01015

2020

Awad, N., Shala, G., Deng, D., Mallik, N., Feurer, M., Eggensperger, K., Biedenkapp, A., Vermetten, D., Wang, H., Doerr, C., Lindauer, M., & Hutter, F. (2020). Squirrel: A Switching Hyperparameter Optimizer.

arxiv.org/abs/2012.08180

Projects

Dynamic Algorithm Configuration

Da die Konfigurationen während der Laufzeit in Abhängigkeit vom aktuellen Zustand des Algorithmus ausgewählt werden sollten, kann es als ein Problem des Reinforcement Learning (RL) betrachtet werden, bei dem ein Agent in jedem Zeitschritt die zu verwendende Konfiguration auf der Grundlage der Leistung im letzten Schritt und des aktuellen Zustands des Algorithmus auswählt. Dies ermöglicht uns einerseits den Einsatz leistungsfähiger RL-Methoden, andererseits bringt RL auch eine Reihe von Herausforderungen mit sich, wie Instabilität, Rauschen und Ineffizienz bei der Abtastung, die bei Anwendungen wie DAC angegangen werden müssen. Daher umfasst die Forschung zu DAC auch die Forschung zu zuverlässigem, interpretierbarem, allgemeinem und schnellem Reinforcement Learning.

Led by: Prof. Dr. Marius Lindauer

Year: 2019

Funding: DFG

Duration: 2019-2023
Dynamic Algorithm Configuration

As configurations should be chosen during runtime depending on the current algorithm state, it can be viewed as a reinforcement learning (RL) problem where at each timestep an agent selects the configuration to use based on the performance in the last step and the current state of the algorithm. This enables us to use powerful RL methods on one hand; on the other, RL also brings a set of challenges like instability, noise and sample inefficiency that need to be addressed in applications such as DAC. Therefore research on DAC also includes research on reliable, interpretable, general and fast reinforcement learning.

Led by: Prof. Dr. Marius Lindauer

Year: 2019

Funding: DFG

Duration: 2019-2023
CoyPu: Cognitive Economy Intelligence Plattform für die Resilienz wirtschaftlicher Ökosysteme

Naturkatastrophen, Pandemien, Finanzkrisen, politische Krisen und Angebotsknappeheiten oder Nachfrageschocks propagieren sich durch offensichtliche und latente Handelsbeziehungen durch das globale ökonomische System. Dies ist eine Konsequenz der kontinuierlichen Globalisierung mit der einhergehenden Arbeitsteilung. Ziel dieses Projektes ist es diese Verbindungen offenzulegen und kaskadierende Risiken vorherzusagen um damit Unternehmen die Möglichkeit einzuräumen vorausschauend agieren zu können.

Led by: Prof. Marius Lindauer and Prof. Maria Esther-Vidal (L3S/LUH)

Team: InfAI, DATEV eg., eccenca GmbH, Implisense GmbH, Deutsches Institut für Wirtschaftsforschung, Leibniz Informationszentrum Technik und Naturwissenschaften, Hamburger Informatik Technologie-Center e.V., Selbstregulierung Informationswirtschaft e.V., Infineo

Year: 2021

Funding: Innovationswettbewerb Künstliche Intelligenz (BMWK)

Duration: 2021-2024
Leibniz AI Academy

The Leibniz AI Academy aims to develop and establish a trans-curricular and interdisciplinary micro-degree program at the Leibniz Universität Hannover (LUH), in which students from different courses of study acquire competencies in the field of Artificial Intelligence

Led by: Prof. Dr. Marius Lindauer, Prof. Dr. Ralph Ewert, Prof. Dr. Johannes Krugel

Year: 2021

Funding: Bundesministerium für Bildung und Forschung (BMBF)

Duration: 2021 - 2024
ERC Starting Grant: Interactive and Explainable Human-Centered AutoML

Trust and interactivity are key factors in the future development and use of automated machine learning (AutoML), supporting developers and researchers in determining powerful task-specific machine learning pipelines, including pre-processing, predictive algorithm, their hyperparameters and--if applicable--the architecture design of deep neural networks. Although AutoML is ready for its prime time after it achieved impressive results in several machine learning (ML) applications and its efficiency improved by several orders of magnitudes in recent years, democratization of machine learning via AutoML is still not achieved. In contrast to previously purely automation-centered approaches, ixAutoML is designed with human users at its heart in several stages. First of all, the foundation of trustful use of AutoML will be based on explanations of its results and processes. Therefore, we aim for: (i) Explaining static effects of design decisions in ML pipelines optimized by state-of-the-art AutoML systems. (ii) Explaining dynamic AutoML policies for temporal aspects of dynamically adapted hyperparameters while ML models are trained. These explanations will be the base for allowing interactions, bringing the best of two worlds together: human intuition and generalization capabilities for complex systems, and efficiency of systematic optimization approaches for AutoML. Concretely, we aim for: (iii) Enabling interactions between humans and AutoML by taking human's latent knowledge into account and learning when to interact. (iv) Building first ixAutoML prototypes and showing its efficiency in the context of Industry 4.0. Perfectly aligned with the EU's AI strategy and recent efforts on interpretability in the ML community, we strongly believe that this timely human-centered ixAutoML will have a substantial impact on the democratization of machine learning.

Led by: Prof. Dr. Marius Lindauer

Team: AutoML

Year: 2022

Funding: EU

Duration: 2022-2027
KISSKI: AI Service Center

The central approach for the KISSKI project is the research on AI methods and their provision with the goal of enabling a highly available AI service center for critical and sensitive infrastructures with a focus on the fields of medicine and energy. Due to their relevance to society as a whole, medicine and the energy industry are among the future fields of application-oriented AI research in Germany. Beyond the technological developments, artificial intelligence (AI) has the potential to make a significant contribution to social progress. This is particularly true in areas where digitization processes are increasingly gaining ground and complexity is high. For both medicine and the energy industry, the pressure to innovate, but also the potential, is immense due to the availability of more and more distributed information based on a multitude of new sensors and actuators. The increasing complexity of the tasks as well as the availability of very large data sets offer a high potential for the application of AI methods in both topics.

Led by: Prof. Dr. Marius Lindauer

Team: AutoML

Year: 2022

Funding: BMBF

Duration: 2022-2025