Unlock the Potential of Your Data
Our world is overflowing with data organized in databases and tables, yet the true power of this data remains largely untapped due to the slow progress and complexity of tabular AI. We are on a mission to break down these barriers. We envision a future where cutting-edge tabular AI is accessible to everyone, turning data into knowledge and insight for everyone.
Our AI research is focused on core areas that we believe to matter most for tabular data.
Handsfree Machine Learning
We know that data science can be tricky and data scientists can be scarce. That's why our research is all about making the best machine learning tools easy for everyone to use. No matter your experience level, we're here to help you make the most of your data.
Integrating your data context
We believe in the power of context when it comes to data. Our research aims to develop AI systems that can understand and integrate the unique context of your data, enhancing the accuracy and relevance of insights derived.
Cutting edge AI capabilities
Our research is at the forefront of AI technology, pushing the boundaries of what is possible with tabular data. We strive to develop innovative solutions that can transform the way you interact with and utilize your data.
State-of-the-art tabular classification in less than a second
We present TabPFN, a trained Transformer that can do supervised classification for small tabular datasets in less than a second, needs no hyperparameter tuning and is competitive with state-of-the-art classification methods.
The Power of Context-Aware Solutions
CAAFE emphasizes the significance of context-aware solutions that can extend the scope of AutoML systems to semantic AutoML. By incorporating domain knowledge into the AutoML process, CAAFE automates feature engineering for tabular datasets, generating semantically meaningful features and explanations of their utility.
The method is not only effective but also interpretable, providing a textual explanation for each generated feature. This makes the automated feature engineering process more transparent, enhancing the interpretability of AI models.
Who we are
We are a group of researchers in Frank Hutter's AutoML Lab at the University of Freiburg working towards our vision of a new generation of algorithms for tabular data.