Ana Neacșu

Lecturer, PhD, Faculty of Electronics, Telecommunications and Information Technology,
University Politehnica of Bucharest

Email: ana_antonia.neacsu@upb.ro
Office: Room 211, UPB Campus Research Center

Research Interests
Degrees and Professional Experience
- Education
- Professional experience
Scientific Activity
- International research projects
- Collaborations
- Affiliation to scientific community

Last update: July 2024

List of publications
- Journal papers
- Conference papers
Academic Activity
- Teaching
- Student supervision

Research Interests

Machine Learning, Deep Learning, Artificial Intelligence
Robustness of Neural Networks and Explainable AI
Automatic Audio Processing
Intelligent Signal Processing of Biomedical data

Degrees and Professional Experience

Education

2019 – 2023: Doctoral Degree
- Double Specialization
  - Electronics Engineering and Telecommunications (“Summa cum Laude” distinction) – University Politehnica of Bucharest, Romania
  - Image and Signal Processing – CentraleSupélec, Université Paris-Saclay, France
- Title: “Robust Deep Learning Methods Inspired from Signal Processing Algorithms”
- Advisors: Corneliu Burilanu, Jean-Christophe Pesquet

2017 – 2019: Master of Science
- Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest, Romania
- Thesis title: “Real-Time Gesture Recognition System”
- in collaboration with Centre de Vision Numérique

2013 – 2017: Bachelor of Science
- Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest, Romania
- Specialization: Telecommunication Systems (TST)

2009 – 2013: Baccalaureate Diploma
- National College “Nicolae Bălcescu”, Brăila, Romania
- Specialization: Mathematics and Computer Science

Professional experience

2024 – present : Lecturer
- Department of Electronic Devices, Circuits and Architectures
- Faculty of Electronics, Telecommunications and Information Technology
- University Politehnica of Bucharest, Romania
2021 – present: Head of Audio Research Department
- Collaboration with MADENN
2016 – present : Scientific Researcher
- Speech and Dialogue Research laboratory
- Research assistant in several national and international research projects
- Coordinator of Robotic Systems Laboratory in CAMPUS Research Center (since 2022)
2019 – 2024 : Teaching Assistant
- Department of Electronic Devices, Circuits and Architectures
- Faculty of Electronics, Telecommunications and Information Technology
- University Politehnica of Bucharest, Romania
2019 – 2022 : Teaching Assistant
- CentraleSupélec
- Université Paris Saclay, France

Scientific Activity

National research projects

2017 – 2020 : Researcher in the project PN-III-P2-2.1-SOL-2016-02-0002, agreement 2SOL/2017, funded by the Romanian Government through UEFISCDI: Intelligent Systems for Video and Audio Analysis – Technologies and Innovative Video Systems for Person Re-identification and Analysis of Dissimulated Behavior (SPIA-VA).
2019 – 2022 : Researcher in the project 425PED/2020, funded by the Romanian Government through UEFISCDI: Embedded Platform For EMG Acquisition with Artificial Intelligence Software Toolkit (EMGP-AI).
2023 : Researcher in the project 450/390126/31.03.2023, MySMIS: 156387, Sistem informatic Integrat de Voice to Text Analytics (VOITA).

International collaborations

OPIS-INRIA – Paris, France
MHUG – Trento, Italy
BM/d – Eindhoven, Netherlands

Affiliation to scientific community

2019 – present : Member of Institute of Electrical and Electronics Engineers (IEEE)
2016 – present : Member of SpeeD Laboratory

Organizer

2019 – present: Member of the organizing committee for the International Conference on Speech Technology and Human-Computer Dialogue – SpeD (“https://sped.pub.ro/“).

List of publications

Journal Papers

S. Juvină, A. Neacșu, J. Rony, J.-C. Pesquet, C.Burileanu, “Training Graph Neural Networks Subject to a Tight Lipschitz Constraint”, in Transactions on Machine Learning Research (TMLR), ISSN: 2835-8856, 2024

A. Neacșu, J-C. Pesquet, V.Vasilescu, C.Burileanu, “ABBA Neural Networks: Coping with Positivity, Expressivity, and Robustness“, in SIAM Journal on Mathematics of Data Science (SIMODS), vol 6, no 3, pages 649-678, doi: 10.1137/23M1589591, 2024
A. Neacșu, J.-C. Pesquet, and C. Burileanu, “EMG-Based Automatic Gesture Recognition Using Lipschitz-Regularized Neural Networks“, in ACM Transactions on Intelligent Systems and Technology (TIST), vol. 5, no. 2, ISSN: 2157-6904, doi: 10.1145/3635159, 2023, ISI WOS:001208775700007.
N. Lassau, S. Ammari, E. Chouzenoux, A. Neacșu et al. “Integrating deep learning CTscan model, biological and clinical variables to predict the severity of COVID-19 patients“, in Nature Communication vol. 12, no. 634, 2021, ISI WOS:000614504800004.

Conference Papers

B. Moroşanu, M. Negru, A. Neacşu, C. Negrescu and C. Paleologu, “Personalized Multi-Track Leveling Algorithm“, in Proc. International Conference on Speech Technology and Human-Computer Dialogue (SpeD), Bucharest, Romania, 2023, pp. 24-29, doi: 10.1109/SpeD59241.2023.10314904.
M. Negru, B. Moroşanu, A. Neacşu, D. Drăghicescu and C. Negrescu, “Automatic Audio Upmixing Based on Source Separation and Ambient Extraction Algorithms“, in Proc. International Conference on Speech Technology and Human-Computer Dialogue (SpeD), Bucharest, Romania, 2023, pp. 12-17, doi: 10.1109/SpeD59241.2023.10314957.
C. Andronache, M. Negru, I. Bădiţoiu, G. Cioroiu, A. Neacșu, and C. Burileanu, “Automatic Gesture Recognition Framework Based on Forearm EMG Activity“, in Proc. 45^th International Conference on Telecommunications and Signal Processing (TSP), Prague, Czech Republic, 2022, pp. 284-288, doi: 10.1109/TSP55681.2022.9851314, ISI WOS:001070846300059.
A. Neacșu, R. Ciubotaru, J.-C. Pesquet, and C. Burileanu, “Design of Robust Complex–Valued Feed- Forward Neural Networks“, in Proc. 30^th European Signal Processing Conference (EUSIPCO), Belgrade, Serbia, 2022, pp. 1596-1600, doi: 10.23919/EUSIPCO55093.2022.9909696, ISI WOS:000918827600313.
A. Neacșu, K. Gupta, J.-C. Pesquet, and C. Burileanu, “Signal Denoising Using a New Class of Robust Neural Networks“, in Proc. of 28^th European Signal Processing Conference (EUSIPCO), Amsterdam, Netherlands, 2021, pp. 1492-1496, doi: 10.23919/Eusipco47968.2020.9287630, ISI WOS:000786144100128.
V. Vasilescu, A. Neacşu, E. Chouzenoux, J.-C. Pesquet, and C. Burileanu, “A Deep Learning Approach For Improved Segmentation Of Lesions Related To Covid-19 Chest CT Scans“, in Proc. IEEE 18th International Symposium on Biomedical Imaging (ISBI), Nice, France, 2021, pp. 635-639, doi: 10.1109/ISBI48211.2021.9434139, ISI WOS:000632622300300.
A. Neacșu, J.-C. Pesquet, and C. Burileanu, “Accuracy-robustness trade-off for positively weighted neural networks“, in Proc. IEEE International Conference on Acoustics and Speech Signal Processing (ICASSP), pp. 8389–8393, Barcelona, Spain, 2020, doi: 10.1109/ICASSP40776.2020.9053803, ISI WOS:000615970408132.
* C. Andronache, M. Negru, A. Neacșu, G. Cioroiu, A. Radoi, and C. Burileanu, “Towards extending real-time EMG-based gesture recognition system” ,in Proc. 43^rd International Conference on Telecommunications and Signal Processing (TSP), Milan, Italy, 2020, pp. 301-304, doi: 10.1109/TSP49548.2020.9163481, ISI WOS:000577106400064. **best paper award**
A. Neacșu, G.Cioroiu, A. Rădoi, C. Burileanu, “Automatic EMG-based Hand Gesture Recognition System using Time-Domain Descriptors and Fully-Connected Neural Networks“, in Proc 42^nd International Conference on Telecommunications and Signal Processing (TSP), Budapest, Hungary, 2019, pp. 232-235 , doi: 10.1109/tsp.2019.8768831, ISI WOS: 000493442800051.
A. Neacșu, C. Burileanu, H. Cucu, “Autonomous System for Performing Dexterous, Human-Level Manipulation Tasks as Response to External Stimuli in Real Time“, in Proc. Future Access Enablers For Ubiquitous and Intelligent Infrastructures (FABULOUS), Bucharest, Romania, 2017, vol. 241, pp. 246-252, doi: 10.1007/978-3-319-92213-3_36, ISI WOS: 000481658200036.

Academic Activity

Teaching

I have a permanent position as a lecturer in the Faculty of Electronics, Telecommunications and Information Technology, National University of Science and Technology POLITEHNICA Bucharest.

Master curricula

2024 – present : Artificial Intelligence II: Deep Neural Networks , BIOSINF Master’s program, 1st year: lectures
2024 – present : Artificial Intelligence III: Advanced deep learning methods, BIOSINF Master’s program, 2nd year: lectures

Bachelor curricula

2024 – present : Microprocessor Architectures 2. Microcontrollers, ETTI, UPB, 2nd year: lectures and labs
2017 – present : Project 2: Applied Electronics, ETTI, UPB, 3rd year
2017 – 2023 : Microcontrollers, ETTI, UPB, 3rd year : labs
2020 – 2022 : ST7 Optimisation, CentraleSupélec, UPS, France, 2nd year : TD
2018 – 2020 : Microprocessor Architectures, ETTI, UPB, 2nd year: labs

Other Teaching activities

Application part of Numerical optimization methods in imaging at International Ph.D. Summer School Mathematics and Machine Learning for Image Analysis, 4-12 June 2024, Bologna, Italy

Student supervision

Graduate students:

Bogdan TUDOR, “Algoritm de generare de watermark bazat pe utilizarea transformatei cosinus discrete“, TAID 2024, ETTI, POLITEHNICA Bucharest
Cătălin CIOCÎRLAN, “Intelligent portrait generation system using Style Transfer algorithms“, SIVA 2024, ETTI, POLITEHNICA Bucharest
Andrei DĂNILĂ, “Robust Deep Learning Algorithms for Deepfake Detection“, SIVA 2024, ETTI, POLITEHNICA Bucharest
Ramona ROTARU, “Algoritmi de învățare automată pentru detecția depresiei pe baza de EEG”, EIM 2024, ETTI, POLITEHNICA Bucharest
Dan CURĂVALE, “Sistem robust și agnostic de procesare a semnalelor EEG cu aplicație în detecția anomaliilor epileptice”, EIM 2024, ETTI, POLITEHNICA Bucharest
Alexandru GUZU, “Dezvoltarea unor metode avansate de analiză automată a semnalelor sEMG” , EIM 2024, ETTI, POLITEHNICA Bucharest
Mihai MUNTEAN, “Sistem pentru detecția activității sportive și monitorizarea utilizatorului”, EIM 2024, ETTI, POLITEHNICA Bucharest
Andrei DĂESCU, “Remote Control Based on sEMG Signals” AM 2023, ETTI, POLITEHNICA Bucharest
Vlad VASILESCU, “Strategii de antrenare pentru rețele neurale convoluționale robuste împotriva atacurilor adversariale“, IISC 2022, ETTI, POLITEHNICA Bucharest
Marian NEGRU, “Sistem automat pentru descrierea contextului dintr-o secvență video“, TAID 2022, ETTI, POLITEHNICA Bucharest
Cristina ANDRONACHE, “Clasificarea inteligentă a topografiilor provenite din semnale EEG”, TAID 2022, ETTI, POLITEHNICA Bucharest
Mădălin COSTEA, “Optimizarea super rezoluției video în timp real”, TAID 2022, ETTI, POLITEHNICA Bucharest