Keynotes
We are excited to announce the following two keynotes.
- Saleema Amershi (Microsoft Research): Challenges in creating responsible and human-centered AI
- Christopher Re (Stanford University): Software 2.0, Foundation Models, Data-Centric AI, and why I’m excited enough to tolerate these buzzwords.
Accepted contributions
We are excited to announce the list of accepted papers, industry talks, and posters.
Tentative Program
| Pacific Time | Eastern Time | Central European | Beijing Time | |
|---|---|---|---|---|
May 16, 2022 | ||||
| 3:00 | 6:00 | 12:00 | 18:00 | Welcome to CAIN |
| Session 1 Quality Assurance | ||||
| 3:30 | 6:30 | 12:30 | 18:30 | What is Software Quality for AI Engineers? Towards a Thinning of the Fog Valentina Golendukhina, Valentina Lenarduzzi, Michael Felderer |
| 3:45 | 6:45 | 12:45 | 18:45 | Exploring ML testing in practice - Lessons learned from an interactive rapid review with Axis Communications Qunying Song, Markus Borg, Emelie Engstrom, Håkan Ardö, Sergio Rico |
| 4:00 | 7:00 | 13:00 | 19:00 | Quality Assurance of Generative Dialog Models in an Evolving Conversational Agent Used for Swedish Language Practice Markus Borg, Johan Bengtsson, Harald Österling, Alexander Hagelborn, Isabella Gagner, Piotr Tomaszewski |
| 4:15 | 7:15 | 13:15 | 19:15 | Discussion |
| 4:30 | 7:30 | 13:30 | 19:30 | Break |
| Session 2 Posters | ||||
| 4:45 | 7:45 | 13:45 | 19:45 | Activity (TBD) |
| 5:15 | 8:15 | 14:15 | 20:15 | MLOps: Five Steps to Guide its Effective Implementation Beatriz Matsui, Denise Goya |
| 5:18 | 8:18 | 14:18 | 20:18 | Towards A Methodological Framework for Production-ready AI-based Software Components Markus Haug, Justus Bogner |
| 5:21 | 8:21 | 14:21 | 20:21 | Preliminary Insights to enable automation of the Software Development Process in Software StartUps. A Investigation Study from the use of Artificial Intelligence and Machine Learning Olimar Borges, Valentina Lenarduzzi, Rafael Prikladnicki |
| 5:24 | 8:24 | 14:24 | 20:24 | Identification of Out-of-Distribution Cases of CNN using Class-Based Surprise Adequacy Mira Marhaba, Ettore Merlo, Foutse Khomh, Giuliano Antoniol |
| 5:27 | 8:27 | 14:27 | 20:27 | Robust Active Learning: Sample-Efficient Training of Robust Deep Learning Models Yuejun GUo, Qiang Hu, Maxime Cordy, Mike Papadakis, Yves Le Traon |
| 5:30 | 8:30 | 14:30 | 20:30 | A New Approach for Machine Learning Security Risk Assessment Jun Yajima, Maki Inui, Takanori Oikawa, Fumiyoshi Kasahara, Ikuya Morikawa, Nobukazu Yoshioka |
| 5:33 | 8:33 | 14:33 | 20:33 | TopSelect: A Topology-based Feature Selection Method for Industrial Machine Learning Hadil Abukwaik, Lefter Sula, Pablo Rodriguez |
| 5:36 | 8:36 | 14:36 | 20:36 | Pynblint: a Static Analyzer for Python Jupyter Notebooks Luigi Quaranta, Fabio Calefato, Filippo Lanubile |
| 5:39 | 8:39 | 14:39 | 20:39 | Traceable Business-to-Safety Analysis Framework for Safety-critical Machine Learning Systems Jati Hiliamsyah Husen, Hironori Washizaki, Hnin Tun, Nobukazu Yoshioka, Hironori Takeuchi, Yoshiaki Fukazawa |
| 5:42 | 8:42 | 14:42 | 20:42 | Structural Causal Models as Boundary Objects in AI System Development Hans-Martin Heyn, Eric Knauss |
| 5:45 | 8:45 | 14:45 | 20:45 | Poster Visits |
| 6:15 | 9:15 | 15:15 | 21:15 | Break |
| Session 3 Training & Learning | ||||
| 6:30 | 9:30 | 15:30 | 21:30 | An Empirical Evaluation of Flow Based Programming in the Machine Learning Deployment Context Andrei Paleyes, Christian Cabrera, Neil D. Lawrence |
| 6:45 | 9:45 | 15:45 | 21:45 | Automatic Checkpointing and Deterministic Training for Deep Learning Xiangzhe Xu, Hongyu Liu, Guanhong Tao, Zhou Xuan, Xiangyu Zhang |
| 7:00 | 10:00 | 16:00 | 22:00 | Influence-Driven Data Poisoning in Graph-Based Semi-Supervised Classifiers Adriano Franci, Maxime Cordy, Martin Gubri, Mike Papadakis, Yves Le Traon |
| 7:15 | 10:15 | 16:15 | 22:15 | Engineering a Platform for Reinforcement Learning Workloads [Industry Talk] Ali Kanso, Kinshuman Patra |
| 7:30 | 10:30 | 16:30 | 22:30 | Method Cards for Prescriptive Machine-Learning Transparency David Adkins, Bilal Alsallakh, Adeel Cheema, Narine Kokhlikyan, Emily McReynolds, Pushkar Mishra, Chavez Procope, Jeremy Sawruk, Erin Wang, Polina Zvyagina |
| 7:45 | 10:45 | 16:45 | 22:45 | Discussion |
| 8:00 | 11:00 | 17:00 | 23:00 | Break |
| 8:15 | 11:15 | 17:15 | 23:15 | Session 4 Keynote Christopher Re: Software 2.0, Foundation Models, Data-Centric AI, and why I’m excited enough to tolerate these buzzwords. |
| 9:15 | 12:15 | 18:15 | 0:15 (+1) | End of Day 1 |
May 17, 2022 | ||||
| Session 5 AI Engineering Practices | ||||
| 3:00 | 6:00 | 12:00 | 18:00 | Towards a Roadmap for Software Engineering for Responsible AI Qinghua Lu, Liming Zhu, Xiwei Xu, Jon Whittle, Zhenchang Xing |
| 3:15 | 6:15 | 12:15 | 18:15 | AI Governance in the System Development Life Cycle: Insights on Responsible Machine Learning Engineering Samuli Laato, Teemu Birkstedt, Matti Mäntymäki, Matti Minkkinen, Tommi Mikkonen |
| 3:30 | 6:30 | 12:30 | 18:30 | The Goldilocks Framework: Towards Selecting the Optimal Approach to Conducting AI Projects Rimma Dzhusupova, Jan Bosch, Helena Holmström Olsson |
| 3:45 | 6:45 | 12:45 | 18:45 | What Is an AI Engineer? An Empirical Analysis of Job Ads in the Netherlands Marcel Meesters, Petra Heck, Alexander Serebrenik |
| 4:00 | 7:00 | 13:00 | 19:00 | Data is about Detail: An Empirical Investigation for Software Systems with NLP at Core Anmol Singhal, Preethu Rose Anish, Pratik Sonar, Smita Ghaisas |
| 4:15 | 7:15 | 13:15 | 19:15 | Discussion |
| 4:30 | 7:30 | 13:30 | 19:30 | Break |
| Session 6 AI Models & Pipelines | ||||
| 4:45 | 7:45 | 13:45 | 19:45 | Practical Insights of Repairing Model Problems on Image Classification [Industry Talk] Akihito Yoshii, Susumu Tokumoto, Fuyuki Ishikawa |
| 5:00 | 8:00 | 14:00 | 20:00 | UDAVA: An Unsupervised Learning Pipeline for Sensor Data Validation in Manufacturing Erik Johannes Husom, Simeon Tverdal, Arda Goknil, Sagar Sen |
| 5:15 | 8:15 | 14:15 | 20:15 | Black-Box Models for Non-Functional Properties of AI Software Systems Daniel Friesel, Olaf Spinczyk |
| 5:30 | 8:30 | 14:30 | 20:30 | Improving Generalizability of ML-enabled Software through Domain Specification Hamed Barzamini, Mona Rahimi, Murtuza Shahzad, Hamed Alhoori |
| 5:45 | 8:45 | 14:45 | 20:45 | Data Sovereignty for AI Pipelines: Lessons Learned from an Industrial Project at Mondragon Corporation Marcel Altendeitering, Julia Pampus, Felix Larrinaga, Jon Legaristi, Falk Howar |
| 6:00 | 9:00 | 15:00 | 21:00 | Discussion |
| 6:15 | 9:15 | 15:15 | 21:15 | Break |
| Session 7 Smells | ||||
| 6:30 | 9:30 | 15:30 | 21:30 | Activity (TBD) |
| 7:00 | 10:00 | 16:00 | 22:00 | Code Smells for Machine Learning Applications Haiyin Zhang, Luís Cruz, Arie van Deursen |
| 7:15 | 10:15 | 16:15 | 22:15 | Data Smells: Categories, Causes and Consequences, and Detection of Suspicious Data in AI-based Systems Harald Foidl, Michael Felderer, Rudolf Ramler |
| 7:30 | 10:30 | 16:30 | 22:30 | Data Smells in Public Datasets Arumoy Shome, Luís Cruz, Arie van Deursen |
| 7:45 | 10:45 | 16:45 | 22:45 | Discussion |
| 8:00 | 11:00 | 17:00 | 23:00 | Break |
| 8:15 | 11:15 | 17:15 | 23:15 | Session 8 Keynote Saleema Amershi: Challenges in creating responsible and human-centered AI |
| 9:15 | 12:15 | 18:15 | 0:15 (+1) | Closing and Awards |
| 9:30 | 12:30 | 18:30 | 0:30 (+1) | End of Day 2 |