Natalia Pavlovskaia

I worked on speech enhancement for a hearing aid, using Tensorflow.

• Target Speaker extraction by the voice sample/direction

• Speech Enhancement (denoising)

• Models Compression (knowledge distillation, NAS, quantization)

I worked on a humanoid robot trained using reinforcement learning to do tasks such as pick and place.

• Perception: autoencoders, contrastive learning, 3D object pose estimation. Robot achieved >90% success rate.

• Sensor fusion: made the system robust to losing the data from 1-2 out of 4 different sensors.

• Dynamics: created a model of the robot's dynamics, which improved the success rate in simulation by 5%.

• Behavioral cloning: trained policies for different tasks with success rates above 90%.

• Engineering: made the real robot 3.3 times faster.

A face recognition system can be fooled by the presentation of a printed photo or replayed video. So I worked on the passive face antispoofing task.

• data collection (academic data, scraping, crowd-sourcing)

• deep learning models implementation and training (using Pytorch)

• moving and testing the trained models in C++ compatible format

• creation of the working demo, including the front-end (JS). My main contribution was the domain adaptation part, which brought an improvement of 10% HTER.