Yannis Assael
yannis assael

Yannis Assael

Greek artificial intelligence research scientist
working at Google DeepMind.
Download Resume

News & Media

International media

29/09/2022. AI is already better at lip reading than we are, Engadget.
13/07/2022. Alumni Profile - Yannis Assael, University of Oxford Inspired Research.
24/05/2022. Una herramienta de inteligencia artificial ayuda a la restauración y atribución de textos antiguos, El País.
13/03/2022. DeepMind’s New AI Helps Restore Damaged Ancient Texts, Wired.
10/03/2022. AI helps historians complete ancient Greek inscriptions damaged over millennia, TechCrunch.
10/03/2022. AI fills the gaps in ancient texts to clear Pericles, The Times.
09/03/2022. Ancient mysteries to be revealed as AI fills in the blanks of damaged manuscripts, The Telegraph.
09/03/2022. DeepMind’s new AI model helps decipher, date, and locate ancient inscriptions, The Verge.
09/03/2022. DeepMind claims its AI can decipher ancient Greek texts from damaged artifacts, VentureBeat.
09/03/2022. AI could decipher gaps in ancient Greek texts, say researchers, The Guardian.
09/03/2022. AI can help historians restore ancient texts from damaged inscriptions, NewScientist.
14/04/2021. Forbes names Wolfson Alumnus on 30 Under 30 list, Wolfson College, University of Oxford.
08/04/2021. Department alumnus Yannis Assael featured in Forbes 30 under 30 list for Europe, Dept. Computer Science, University of Oxford.
07/04/2021. 30 Under 30 Science & Healthcare, Forbes.
07/04/2021. 30 Under 30 Europe 2021: Changing The World From Artificial Intelligence To Covid-19, Forbes.
10/11/2020. DeepMind Says Impact of New End-to-End Machine Dubbing Tech May Be Widespread, Slator.
01/08/2020. How AI helps historians solve ancient puzzles, Financial Times.
18/12/2019. DeepMind and Google recreate former NFL linebacker Tim Shaw’s voice using AI, VentureBeat.
18/12/2019. Using WaveNet technology to reunite speech-impaired users with their original voices, DeepMind.
04/11/2019. Oracle of AIsolves classic ancient conundrums, The Times.
31/10/2019. Restoring ancient Greek inscriptions using AI deep learning, University of Oxford.
22/10/2019. New research uses deep learning to restore ancient Greek texts, Dept. Computer Science, University of Oxford.
21/10/2019. Google’s DeepMind AI Beats Humans Again—This Time By Deciphering Ancient Greek Text, Observer.
18/10/2019. DeepMind AI beats humans at deciphering damaged ancient Greek tablets, NewScientist.
15/10/2019. Restoring ancient text using deep learning: a case study on Greek epigraphy, DeepMind.
01/10/2019. Al is helping scholars restore ancient Greek texts on stone tablets, TechCrunch.
31/07/2018. Lip-reading artificial intelligence could help the deaf—or spies, Science.
10/11/2016. New software could give lip-reading powers to everyone, Mashable.
08/11/2016. AI that lip-reads 'better than humans', BBC News.
07/11/2016. Can deep learning help solve lip reading?, The Verge.
24/02/2016. Business Insider, Business Insider.
20/02/2016. Can you solve the '100 hat riddle' used by Google in job interviews?, Independent.
19/02/2016. Google's Artificial Intelligence Solves Riddle Used In Google Interviews, IFL Science.
17/02/2016. AI solves 100-hat puzzle used in Google job interviews, NewScientist.

Greek media

16/05/2022. Επετειακή εκδήλωση εορτασμού των 70 χρόνων λειτουργίας, ΙΚΥ.
23/04/2022. Η «βαθιά» ανάγνωση των επιγραφών, Το Βήμα.
28/03/2022. Όμικρον Τρία - Τεχνητή Νοημοσύνη για τον Πολιτισμό, ΕΡΤ3.
14/03/2022. «Ιθάκη»: Ανακαλύπτοντας την αρχαία Ελλάδα μέσα από την Τεχνητή Νοημοσύνη, Υπουργείο Πολιτισμού και Αθλητισμού.
10/03/2022. DeepMind: Σύστημα τεχνητής νοημοσύνης αποκατέστησε για πρώτη φορά κείμενα σε αρχαιοελληνικές επιγραφές, ΕΡΤ.
09/03/2022. Tεχνητή νοημοσύνη της Google αποκρυπτογράφησε αρχαία Ελληνικά κείμενα, Το Βήμα.
09/03/2022. «Ιθάκη» εκεί που η τεχνητή νοημοσύνη συναντά την Αρχαία Ελλάδα, Καθημερινή.
09/03/2022. Έλληνες κατασκεύασαν την «Ιθάκη» και αποκαθιστούν κείμενα σε αρχαιοελληνικές επιγραφές, Μακεδονία.
10/10/2021. Με νέα ματιά - Τεχνητή Νοημοσύνη, ΕΡΤ1.
26/09/2021. Συνδέοντας την Τεχνητή Νοημοσύνη με τον Πολιτισμό, ΒΗΜΑgazino Πρόσωπα.
30/04/2021. Τέσσερις νέοι Ελληνες στη λίστα «Forbes», Το Βήμα.
16/04/2021. Όμικρον Τρία - Τεχνητή Νοημοσύνη, ΕΡΤ3.
15/04/2021. Γιάννης Ασσαέλ στο newsbeast.gr: Νιώθω περηφάνια ως Έλληνας που βρίσκομαι στη λίστα του Forbes, newsbeast.gr.
14/04/2021. Forbes : Αυτοί είναι οι Έλληνες που κατάφεραν να μπουν στη λίστα με τους πιο επιτυχημένους νέους της Ευρώπης, Τα Νέα.
13/04/2021. Τέσσερις Έλληνες επιστήμονες κάτω των 30 ετών στη λίστα του Forbes, LiFO.gr.
12/04/2021. Τρεις Ελληνες επιστήμονες κάτω των 30 ετών στη λίστα του Forbes, Καθημερινή.
11/04/2021. Ποιοι είναι οι Έλληνες που μπήκαν στην ευρωπαϊκή λίστα του Forbes "30 κάτω των 30" 2021, Capital.gr.
08/04/2021. Γιάννης Ασσαέλ: Ποιος είναι ο Θεσσαλονικιός που φιγουράρει στη λίστα του Forbes Under 30, Glow.
12/11/2020. Όταν η τεχνητή νοημοσύνη «μεταφράζει» τον ήχο, προς όφελος της εκπαίδευσης, Αθηναϊκό - Μακεδονικό Πρακτορείο Ειδήσεων.
08/09/2020. Δρ Γιάννης Ασσαέλ ‘08: Εντυπώσεις από την Τελετή Αποφοίτησης του High School 2020, Anatolia College.
19/08/2020. Γιάννης Ασσαέλ: «Φυσικά πιστεύω πως μπορούμε ν' αλλάξουμε τον κόσμο», Αθηναϊκό - Μακεδονικό Πρακτορείο Ειδήσεων.
20/10/2019. «Πυθία»: Ο γόνος της Google και Ελληνα ερευνητή που «διαβάζει» μισοκατεστραμμένες αρχαίες επιγραφές, Kαθημερινή.
01/08/2018. Θεσσαλονικιός πίσω από σύστημα τεχνητής νοημοσύνης της Google που «διαβάζει» τα χείλη, Parallaxi.
01/08/2018. Νέο πρόγραμμα τεχνητής νοημοσύνης δημιούργησε ομάδα ερευνητών με επικεφαλής Ελληνα, Kαθημερινή.
23/11/2017. 6o Μαθητικό Συνέδριο Επιστήμης & Τεχνολογίας ACSTAC 2017, Kαθημερινή.
22/11/2017. Διακεκριμένοι επιστήμονες σε μαθητικό συνέδριο στη Θεσσαλονίκη, TyposThes.
28/11/2016. Ο Γιάννης Ασσαέλ μιλά για το ριζοσπαστικό σύστημα τεχνητής νοημοσύνης που διαβάζει τα ανθρώπινα χείλη, LiFO.gr.
17/11/2016. LipNet - Τώρα με την Άννα Μπουσδούκου, ΣΚΑΙ.
09/11/2016. Λογισμικό διαβάζει τα χείλη των ανθρώπων, Το Βήμα.
01/03/2016. Αυτός είναι ο γρίφος των 100 καπέλων που σε βάζουν να λύσεις η Google και η Goldman Sachs αν τους ζητήσεις δουλειά, LiFO.gr.
08/12/2010. Η HTC βράβευσε τους νικητές του GRAND - Greek Android Developers Challenge, PC Magazine.

Curriculum Vitae

Education

DPhil in Machine Learning

2015 - 2019 University of Oxford
  • Supervisor: Shimon Whiteson, Nando de Freitas
  • Oxford – Google DeepMind Graduate Scholarship

MRes in Machine Learning

Distinction, 81.8%, 2nd best overall performance

2014 - 2015 Imperial College
  • Supervisor: Marc P. Deisenroth
  • HiPEDS Scholarship (EPSRC)

MSc in Computer Science

Distinction, 83.5%, best overall performance

2013 - 2014 University of Oxford
  • Supervisor: Nando de Freitas
  • Tony Hoare Prize for best overall performance
    in the MSc in Computer Science 2014
  • Hellenic State Scholarships Foundation (IKY)

Security in IT Summer School

Distinction A, ECTS 7.5

2011 Engineering College of Copenhagen (IHK)

Diploma of Applied Informatics

83.7%, 4-Years

2008 - 2013 University of Macedonia
  • Supervisor: Konstantinos G. Margaritis

High School

17.8 / 20

2002 - 2008 Anatolia College, Greece

Scholarships

Oxford - Google DeepMind Graduate Scholarship

full scholarship for DPhil in Machine Learning, 2015 - 2018.

Imperial College London

full scholarship for MRes program of "HiPEDS", 2014 - 2015.

Hellenic State Scholarships Foundation

full scholarship for MSc in University of Oxford, 2013-2014.

Advanced Training in Open Source Multimedia Instruments

Blender 3D Design, EU Media Program Scholarship, 2011 - Sofia, Bulgaria.

Training in Open Source Multimedia Instruments

Blender 3D Design, EU Media Program Scholarship, 2010 - Sofia, Bulgaria.

Honors & Awards

Featured on the cover of Nature journal

2022 Nature Publishing Group

Best paper award

2022 Plastic and Reconstructive Surgery - Global Open

Featured in top scholars of the Hellenic State Scholarships Foundation

2022 Event for the 70-year anniversary

Forbes 30 Under 30 Europe

2021 Science & Healthcare

Distinction in MRes in Machine Learning

2015 Imperial College London

Tony Hoare Prize for "Best overall performance in the MSc in Computer Science"

2014 University of Oxford

Distinction in MSc in Computer Science

2014 University of Oxford

1st place in University of Oxford: Machine Learning Competition

Michaelmas 2013 University of Oxford

Distinction in Security in IT

2011 Copenhagen University College of Engineering (IHK)

3rd place in the HTC "Grand" Competition

2010 HTC
  • A competition for Google Android developers, under the "airquality.gr" application.

Certifications

Certified Scrum Master (CSM®)

Scrum Alliance, 2021.

Download Resume

Publications & Patents

Publications

1.Yannis Assael, Thea Sommerschield, Brendan Shillingford, Mahyar Bordbar, John Pavlopoulos, Marita Chatzipanagiotou, Ion Androutsopoulos, Jonathan Prag, Nando de Freitas. Restoring and attributing ancient texts using deep neural networks, Nature, 603, 7900, 280-283, Nature Publishing Group, 2022.
2.Ziga Avsec, Vikram Agarwal, Daniel Visentin, Joseph R Ledsam, Agnieszka Grabska-Barwinska, Kyle R Taylor, Yannis Assael, John Jumper, Pushmeet Kohli, David R Kelley. Effective gene expression prediction from sequence by integrating long-range interactions, Nature Methods, 18, 10, 1196-1203, Nature Publishing Group, 2021.
3.Brendan Shillingford, Yannis Assael, Misha Denil, . Interactive decoding of words from visual speech recognition models, arXiv preprint arXiv:2107.00692, 2021.
4.Angelos Mantelakis, Yannis Assael, Parviz Sorooshian, Ankur Khajuria, . Machine learning demonstrates high accuracy for disease diagnosis and prognosis in plastic surgery, Plastic and Reconstructive Surgery - Global Open, 9, 6, e3638, LWW, 2021.
5.Yi Yang, Brendan Shillingford, Yannis Assael, Miaosen Wang, Wendi Liu, Yutian Chen, Yu Zhang, Eren Sezener, Luis C Cobo, Misha Denil, Yusuf Aytar, Nando de Freitas. Large-scale multilingual audio visual dubbing, arXiv preprint arXiv:2011.03530, 2020.
6.Takaki Makino, Hank Liao, Yannis Assael, Brendan Shillingford, Basilio Garcia, Otavio Braga, Olivier Siohan. Recurrent neural network transducer for audio-visual speech recognition, IEEE automatic speech recognition and understanding workshop (ASRU), 905-912, IEEE, 2019.
7.Yannis Assael, Thea Sommerschield, Jonathan Prag, . Restoring ancient text using deep learning: a case study on Greek epigraphy, Empirical Methods in Natural Language Processing, 2019.
8.Brendan Shillingford, Yannis Assael, Matthew W. Hoffman, Thomas Paine, Cían Hughes, Utsav Prabhu, Hank Liao, Hasim Sak, Kanishka Rao, Lorrayne Bennett, . Large-scale visual speech recognition, Interspeech, 4135-4139, 2019.
9.Archit Gupta, Brendan Shillingford, Yannis Assael, Thomas C Walters, . Speech bandwidth extension with WaveNet, IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2019.
10.Yutian Chen, Yannis Assael, Brendan Shillingford, David Budden, Scott Reed, Heiga Zen, Quan Wang, Luis C Cobo, Andrew Trask, Ben Laurie, . Sample Efficient Adaptive Text-to-Speech, arXiv preprint arXiv:1809.10460, 2018.
11.Nikolaos D Goumagias, Dimitrios Hristu-Varsakelis, Yannis M Assael, . Using deep Q-learning to understand the tax evasion behavior of risk-averse firms, Expert Systems with Applications, 101, 258-270, Pergamon, 2018.
12.Rui Ponte Costa, Yannis Assael, Brendan Shillingford, Nando de Freitas, Tim Vogels, . Cortical microcircuits as gated-recurrent neural networks, Advances in Neural Information Processing Systems, 272-283, 2017.
13.Yannis M Assael, Konstantinos D Antoniadis, Marc J Assael, . From analog timers to the era of machine learning: The case of the transient hot-wire technique, AIP Conference Proceedings, 1866, 1, 20001, AIP Publishing, 2017.
14.Yannis M Assael, Brendan Shillingford, Shimon Whiteson, Nando de Freitas, . LipNet: End-to-End Sentence-level Lipreading, GPU Technology Conference, 2017, 2016.
15.Hossam Mossalam, Yannis M Assael, Diederik M Roijers, Shimon Whiteson, . Multi-Objective Deep Reinforcement Learning, NIPS Deep Reinforcement Learning Workshop, 2016.
16.Jakob N Foerster, Yannis M Assael, Nando de Freitas, Shimon Whiteson, . Learning to Communicate with Deep Multi-Agent Reinforcement Learning, Advances in Neural Information Processing Systems, 2016.
17.Jakob N Foerster, Yannis M Assael, Nando de Freitas, Shimon Whiteson, . Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, IJCAI Workshop, 2016.
18.John-Alexander M. Assael, Niklas Wahlström, Thomas B Schön, Marc Peter Deisenroth, . Data-Efficient Learning of Feedback Policies from Image Pixels using Deep Dynamical Models, NIPS Deep Reinforcement Learning Workshop, 2015.
19.John-Alexander M. Assael, Marc Peter Deisenroth, . From Pixels to Torques: Policy Learning using Deep Dynamical Convolutional Networks, Imperial College London, 2015.
20.Marc J Assael, Konstantinos D Antoniadis, Ifigeneia N Metaxa, Sofia K Mylona, John-Alexander M. Assael, Jiangtao Wu, Miaomiao Hu, . A Novel Portable Absolute Transient Hot-Wire Instrument for the Measurement of the Thermal Conductivity of Solids, International Journal of Thermophysics, 36, 11-Oct, 3083-3105, Springer US, 2015.
21.Charalampos S. Kouzinopoulos, John-Alexander M. Assael, Themistoklis K. Pyrgiotis, Konstantinos G. Margaritis, . A Hybrid Parallel Implementation of the Aho-Corasick and Wu-Manber Algorithms Using NVIDIA CUDA and MPI Evaluated on a Biological Sequence Database, International Journal on Artificial Intelligence Tools, 24, 1, 1540001, World Scientific, 2015.
22.John-Alexander M. Assael, Ziyu Wang, Bobak Shahriari, Nando de Freitas, . Heteroscedastic Treed Bayesian Optimisation, NIPS Workshop on Bayesian Optimization, 2014.
23.John-Alexander M. Assael, Konstantinos Margaritis, . String Matching on hybrid parallel architectures, an approach using MPI and NVIDIA CUDA, http://www.yannisassael.com/publications/assael_uom_dissertation.pdf, University of Macedonia, 2013.
24.MJ Assael, Y. M. Assael, ML Huber, RA Perkins, Y Takata, . Correlation of the Thermal Conductivity of Normal and Parahydrogen from the Triple Point to 1000 K and up to 100 MPa, Journal of Physical and Chemical Reference Data, 40, 3, 033101-033101-13, American Institute of Physics, 2011.
25.M.J. Assael, K.E. Kakosimos, K.D. Antoniadis, John-Alexander M. Assael, . Applying Thermal Comfort Indices to Investigate Aspects of the Climate Change in Greece, International Review of Chemical Engineering, 2, 204-209, Praise Worthy Prize, 2010.
26.M.J. Assael, K.E. Kakosimos, A. Alexandridis, John-Alexander M. Assael, . Study and Application of Discomfort Indices in Greece, 7th National Scientific Chemical Engineering Conference, Patra, Proceedings, 2009.
27.M. J. Assael, John-Alexander M. Assael, K. Kakosimos, . Discomfort Indices - www.airquality.gr, Technographima (Journal of the Technical Chamber of Greece), 362, 13-Dec, Technical Chamber of Greece, 2008.

Patents

1.Ioannis Alexandros AssaelBrendan Shillingford. Training video data generation neural networks using video frame embeddings, WO Patent 053523A1, 2022.
2.Brendan Shillingford, Ioannis Alexandros Assael, Joao Ferdinando Gomes de Freitas. Cross-modal sequence distillation, US Patent 11250838, 2022.
3.Yutian Chen, Scott Ellison Reed, Aaron Gerard Antonius van den Oord, Oriol Vinyals, Heiga Zen, Ioannis Alexandros Assael, Brendan Shillingford, Joao Ferdinando Gomes de Freitas. Sample-efficient adaptive text-to-speech, US Patent App. 17/061, 437, 2021.
4.Ioannis Alexandros Assael, Thomas Chadwick Walters, Archit Gupta, Brendan Shillingford. Bandwidth extension of incoming data using neural networks, WO Patent 221846A1, 2020.
5.Jakob Nicolaus Foerster, Ioannis Alexandros Assael. Resolving time-delays using generative models, US Patent 0369946A1, 2019.
6.Brendan Shillingford, Ioannis Alexandros Assael, Joao Ferdinando Gomes De Freitas. Visual speech recognition by phoneme prediction, WO Patent 219968A1, 2019.

Thesis

DPhil ThesisDeep learning for communication: emergence, recognition and synthesis, University of Oxford, 2019. (link)
MRes ThesisFrom pixels to torques: policy learning using deep dynamical convolutional networks, Imperial College London, 2015.
MSc ThesisBayesian Optimization for Heteroscedastic Functions, University of Oxford, 2014. (link)
Bachelor ThesisString matching on hybrid parallel architectures, an approach using MPI and Nvidia CUDA, University of Macedonia, 2013.
Download Resume

Contact