Alex Graves (computer scientist)

Graves earned his Bachelor of Science degree in Theoretical Physics from the University of Edinburgh^[when?] and a PhD in artificial intelligence from the Technical University of Munich supervised by Jürgen Schmidhuber at the Dalle Molle Institute for Artificial Intelligence Research.^[2][3]

After his PhD, Graves was postdoc working with Schmidhuber at the Technical University of Munich and Geoffrey Hinton^[4] at the University of Toronto.

At the Dalle Molle Institute for Artificial Intelligence Research, Graves trained long short-term memory (LSTM) neural networks by a novel method called connectionist temporal classification (CTC).^[5] This method outperformed traditional speech recognition models in certain applications.^[6] In 2009, his CTC-trained LSTM was the first recurrent neural network (RNN) to win pattern recognition contests, winning several competitions in connected handwriting recognition.^[7][8] Google uses CTC-trained LSTM for speech recognition on the smartphone.^[9][10]

Graves is also the creator of neural Turing machines^[11] and the closely related differentiable neural computer.^[12][13] In 2023, he wrote the paper Bayesian Flow Networks.^[14] He currently works as a Staff Research Scientist at InstaDeep.^[15][16] He was previously a research scientist at Google DeepMind in London.^[17]