Hasso-Plattner-Institut
Prof. Dr. Holger Giese
 

Christian Medeiros Adriano (Chris)

Phone: +49 331 5509 319, Office: Building A, Room A-2.7

E-Mail: christian.adriano [at] hpi.de or [at] gmail.com

My: Linked-in and GitHub

NewsResearchTeachingPublications

 


News

>2024.11.11|Accepted workshop at ICSE25. Together with my colleagues Sona (HPI), Daiki (IBM Japan), and Rúben (Krems University - Austria), I am co-organizing the NSE-Neuro-Symbolic Software Engineering  Workshop at the International Conference for Software Engineering - ICSE25, Ottawa, Canada, please check the Call for Papers
>2024.09.20|Accepted paper at AI4AS-ACSOS24. Co-authored with Sona Ghahremani and Prof. Holger Giese. Principled Transfer Learning for Autonomic Systems: A Neuro-Symbolic Vision. presented at 2nd International Workshop on Artificial Intelligence for Autonomous computing Systems (AI4AS 2024)
>2024.09.13|Accepted paper at NFM-ECML24. Co-authored with Til Schniese and Prof. Holger Giese. Leveraging Cross-Snapshot Attention for Identifying Graph Propagation Patterns in Dynamic Real-World Networks presented at The 12th Workshop on New Frontiers in Mining Complex Patterns, ECML - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
>2024.05.08|Accepted paper at ICPE24. Co-authored with Iqra Zafar and Prof. Holger Giese. STIGS: Spatio-Temporal Interference Graph Simulator for Self-Configurable Multi-Tenant Cloud Systems presented at The International Conference ICPE '24 Companion: Companion of the 15th ACM/SPEC International Conference on Performance Engineering
>2023.03.10|Contributed Talk (with Prof. Giese)AI in Software Engineering, at Adesso, Cologne, Germany, slides 
>2022.03.03|Accepted paper at MODELS22: Tool Support for the Teaching of State-Based Behavior ModelingConference: ACM / IEEE 25th International Conference on Model Driven Engineering Languages and Systems (MODELS) 2022 - the 18th Educators Symposium, Montréal, Canada
>2022.07.22|Talk: Neuro-Symbolic to the rescue: Rashomon Effect, Lord's Paradox, Collider Bias and other Insidious Phenomena (and Creatures), at the System Analysis & Modeling Department, HPI, Potsdam, Germany
>2021.10.27|Invited Talk: Too Big to Fail - Building Robust Intelligent Systems with Causal Machine Learning, at the HPI Research School Retreat, Potsdam, Germany, slides
>2021.07.14|Talk: Towards more Reliable Machine Learning-Based Systems - The Need for Methods to Discover Model Invariants, at the HPI Research School Retreat, Potsdam, Germany
>2020.10.24|Talk: Causal and Sequential Decision Models of Software Fault Understanding,  at the HPI Research School Retreat, Potsdam, Germany
>2020.03.03|Accepted paper on Bayesian modeling! Collective Risk Minimization via a Bayesian Model for Statistical Software Testing, Conference: SEAMS-2020, South Korea
>2019.06.19|Invited Talk: Enabling a crowd of programmers to work in parallel to identify, explain, and fix software bugs, Location: Meeting of the German Computer Science Research Training Groups 2019, in Dagstuhl
>2019.04.09|Invited Talk:Tackling the Perfect Fault Understanding Assumption with One Thousand Programmers in the Loop, Location: FutureSoc Symposium
>2018.07.18|Accepted paper: Microtasking Fault Localization, Conference: Doctoral Symposium at the Empirical Software Engineering Conference.
>2018.07.02|Talk: Crowdsourcing the localization and fixing of software faults, Location: Software Engineering Group at the Humbolt University.
>2018.06.15|Accepted paper: Learning Utility-changes for Rule-based Adaptation of Dynamic Architectures - Current and Future work, Conference:  ICAC-2018
>2018.03.09|Talk: Can a crowd identify the cause of a software failure and suggest valid bug fixes?, Location: University of Cape Town

Research Projects

I build causal models to predict and explain task outcomes, particularly tasks that involve subjective information (opinions, knowledge). The models that I build concern three topics: (1) how to aggregate conflicting information (opinions, knowledge), (2) decide if more information is necessary for making a decision, and (3) who should ideally provide the information. I am particularly interested in the cases where software engineers and autonomous agents are both providers and consumers of information. More recently, I have been investigating these concerns under the topic of transfer learning and combining symbolic and learning-based methods (neuro-symbolic) to determine how fragments of information/knowledge can be reused, updated, extended, or forgotten.

Topics

  • Theory: Decision Theory, Voting, Utility theory, Prospect theory, and Measurement theory.
  • Probabilistic Models: Causal Inference, Reinforcement Learning, Multi-Armed Bandits, Bayesian Inference, Markov Models, and Bayesian Optimization.
  • Empirical research: Quasi-experimental and Observational studies, Survival analysis, Analysis of covariance, Time series, and Sensitivity analysis.
  • Application domains: Crowdsourcing, Software Debugging, Self-Adaptive Systems, Traffic-Signal Control, Multi-Agent Systems.
  • Tools/Languages: R, Python, Java, Scala, Rust.

Projects

I am grateful to have the opportunity to work with many brilliant graduate and undergraduate students in individual and group research projects. Follow some of recent research projects that I have coordinated:

Publications

My Google Scholar - link