AI Research
Research Framework
Advancing AI research
We aim to advance knowledge in the field of artificial intelligence (AI) with interdisciplinary research, open-access publications, and our open-source code.
Empowering intelligent solutions
We identify applications for machine learning and develop algorithms and systems that make SAP solutions more efficient, scalable, and transparent.
Fostering collaboration
Partnering with leading research institutes and universities, our academic programs help young researchers apply machine learning in an industrial context.
Machine learning
SAP connects academics and industry experts to expand knowledge about machine learning. With access to insights from SAP customers, we also work with development teams to harness the power of machine learning within SAP products.
Research Areas
Learning with minimal supervision
While supervised learning requires large annotated datasets for model training, machine learning with minimal supervision uses unlabeled data and requires minimal human intervention. Minimal supervision approaches, such as semi-supervised, self-supervised, active learning, and uncertainty modeling, use other learning frameworks to improve accuracy. These methods are relevant where manual labeling is time-intensive or costly.
Few-shot learning
We develop few-shot learning approaches that work when limited data is available for model training. Our approaches include using multimodal data that leverages both images and text, as well as cross-modal hallucination and meta-learning methods. Few-shot learning approaches are relevant in all business scenarios with limited training data, such as the classification of new products in online product catalogs.
Visual question answering
Our integrative visual question answering (VQA) models improve the detection of fine-grained information and VQA evaluation metrics, enabling computers to answer questions about an image using natural language. The application of VQA models spans various industries and ranges from integration in intelligent chatbots to ticketing and invoice processing systems, as well as intelligent information retrieval for disease diagnoses.
Efficient deep learning
State-of-the-art deep learning models involve extensive computational costs and expensive hardware. We develop novel approaches to resource-efficient deep learning, including assessing model complexity, resource-efficient networks, quantization, pruning, and knowledge distillation. Efficient deep learning approaches in industrial applications help minimize monetary cost, power consumption, inference time, and environmental impact.
Privacy and fairness
We develop approaches to machine learning that impose privacy and fairness constraints, such as differential privacy and federated learning, as well as multitask learning. This enables institutions to use generalized diagnostic or prediction models without risking individuals’ privacy. Developing algorithms with fairness constraints also helps mitigate unconscious bias in applications such as credit scoring, bank loans, or résumé matching.
Lifelong learning
We investigate lifelong learning approaches that enable machine learning models to learn the way humans do, incrementally and by utilizing previous knowledge to learn new tasks. Our methods include class-incremental approaches, continual domain adaptation, neuroplasticity, and adaptive capacity expansion. Machine learning models that learn continuously are relevant in all situations where it is not possible to keep historical training data.
Sentiment analysis
Understanding the sentiment and opinions expressed in natural language is a key challenge in natural language processing. We work on novel approaches for sentiment analysis that focus on neural word embedding and attention-based methods. Sentiment analysis is relevant for business sectors, such as telecommunications, banking, insurance, and e-commerce, where opinions about products and services must be analyzed and acted on.
Interpretable machine learning
Machine learning models and algorithms have reached an outstanding level of sophistication, making it challenging to explain predictions. Interpretable machine learning approaches, such as self-supervised and meta-supervised learning and curiosity-driven models, enable the discovery of patterns in data. This is crucial in business applications as it provides transparency and helps explain the underlying reasons for proposed outputs.
Information extraction
Extracting information from structured and unstructured documents is challenging for natural language processing (NLP), and finding labeled data to train machine learning models is difficult. Our research focuses on sequential and two-dimensional approaches for structured documents, combines elements from NLP with computer vision, and incorporates unsupervised and weakly-supervised models using transfer learning into information extraction pipelines.
Publications
Learn about our current research projects and get the latest news on machine learning and artificial intelligence from experts and thought leaders around the world.
Academic Programs
Master's thesis program
Explore machine learning challenges faced by SAP customers and product teams and gain hands-on experience working as part of an industrial research team.
PhD program
Work with rich datasets to find machine learning-based solutions to real-world problems in close collaboration with our global network of research partners.
Visiting scholar program
Expand existing research areas in machine learning and establish new ones with access to rich datasets and business use cases from SAP product teams.
Blogs
Trends in human-machine collaborative learning
Learn about emerging approaches to capitalizing on the distinctive strengths of humans and machines combined in collaborative ways to maximize potential.
Learning to remember with continual learning
Find out how machine learning research yields insights into continual learning to make AI more efficient for your business, increasing profit and ROI.
Deep few-shot learning from a multimodal perspective
Find out why incorporating data from other modalities during the few-shot model training may be a promising step toward solving the Big Data challenge.