PySAD
An open-source Python library that finds unusual events in streaming data, with models that learn from one sample at a time.
Selim FiratYilmaz
I work on representation learning for large, real-world data: foundation models for transactions, compression for images sent over noisy wireless channels, anomaly detection on data streams, and language understanding across 100 languages. The thread is one idea. A good representation is compact, private when it must be, and still useful when the data is noisy or changing.
Manager, Machine Learning ResearchAmerican Express, London
Selected work
Kaira
A PyTorch toolkit to build and test communication systems that use neural networks, from the channel model to the encoder and decoder.
Wireless image delivery with diffusion models
A diffusion model at the receiver repairs images that arrive over noisy wireless channels. Work done at Huawei R&D.
Sentiment analysis in 100 languages
One model for 100 languages that gives each text one or more sentiment labels and gives rare labels more weight in training.
Private collaborative edge inference
Edge devices send their predictions together in one wireless transmission, which saves bandwidth and keeps each model private.
Education
- PhDElectrical and Electronic EngineeringImperial College LondonFocus: machine learning for communication and compression
- MSElectrical and Electronic EngineeringBilkent University
- BSComputer ScienceBilkent University