ENIAC20 First International Workshop on ENgineering Intelligent Applications' Code
Artificial Intelligence is becoming a mainstream concern in everyday software construction. Driven by appealing success stories such as autonomous vehicles, cloud-based intelligent services (e.g., Google Translate), intelligent health-related mobile apps, etc., more and more software companies intend to leverage AI techniques in their products. However, there is a large gap between the Programming/Software Engineering field and the Artificial Intelligence field. Since the late eighties, these subfields of Computer Science have matured independently: they form separated research communities, they have given rise to separate labs in computer science departments and they often comprise different profiles or specialisations in Computer Science Master programmes. The latest insights in Artificial Intelligence and Programming/Software Engineering are therefore not always compatible and/or understood by practitioners. “Adding intelligent behaviour” to a large modern software system is therefore currently more a craft than a solid engineering domain.
Whereas AI in the late 20th century mainly focussed on symbolic techniques (NLP, knowledge representation, concept learning, genetic programming, rule-based languages, expert systems, …), most contemporary AI (and especially Machine Learning) research is mostly subsymbolic (i.e., numerical) in nature. From a programming perspective, most contemporary AI/ML research is little more than loops operating on arrays storing floating point numbers. This trend exacerbates the aforementioned chasm between the Artificial Intelligence domain and the Programming/Software Engineering domain.
New technologies are needed to reconcile the results of both domains in a systematic and well-founded manner. This workshop seeks to solve this problem. Because it is a
Call for Papers
Relevant topics for technical papers include, but are not limited to:
- Differentiable programming frameworks and languages
- DSLs, libraries, and middleware for integrating AI/ML techniques in software systems
- Declarative languages for knowledge representation
- Programming languages and paradigms for implementing AI/ML techniques
- Requirements analysis for designing and architecting AI/ML-intensive systems
- Architectures for online machine learning and real-time model serving
- Design patterns, best practices, and metrics for ensuring the quality of AI/ML-intensive systems
- Language and tool support for implementing, testing, debugging, verifying, and validating AI/ML-intensive systems
- Language and tool support for model management, evolution, and deployment
- Integration of AI/ML workflows (data collection, cleaning, labeling, feature engineering, model training, model evaluation, model deployment, …) in software engineering processes (e.g., MLOps, …)
In addition to purely academic papers about the above topics, we also solicit experience reports, extensions of and case studies about the integration of state-of-the-art AI/ML platforms:
- AI/ML languages and model representations: CLIPS, Datalog, Drools, Prolog, ONNX, …
- AI/ML frameworks: Azure ML, Caffe, PyTorch, TensorFlow, Scikit-learn, …
- Distributed AI/ML and real-time model serving frameworks: Coral, CoreML, ML Kit, Spark ML, …
The workshop accepts two kinds of contributions:
- 6-page technical papers and experience reports in ACM format to be reviewed by at least three members of the program committee. Accepted submissions will be included in the official workshop post-proceedings. The deadline for these submissions is January 15th.
- 1-page presentation abstracts which, when accepted, will be made available informally on the website. The deadline for these submissions is February 1st.
All submissions should provide unpublished and original work that has not been previously accepted for publication nor concurrently submitted for review in another workshop, conference, journal, or book. If the submission is accepted, at least one author must attend the workshop and present the paper in order to include the paper in the proceedings.
If you have any questions, or wonder whether your submission is in scope, please do not hesitate to contact the PC chair.