Applied Deep Learning
Specialization in the core of AI. Tensor mathematics, Transformer architecture, and efficient training of foundational models.
Deep Learning Neural Networks Research
Computational Foundations of Artificial Intelligence
Rigorous training for those seeking to open the “black box”. From gradient calculation to implementing cutting-edge architectures, this program connects mathematical theory with engineering practice in PyTorch/JAX.
Technical Deep Dive
- Mathematics for Deep Learning: Advanced linear algebra, multivariable calculus, and stochastic optimization.
- Modern Neural Architectures: Implementing Transformers (Attention is All You Need), Diffusion Models, and Autoencoders from scratch.
- Efficient Fine-Tuning: Parameter adaptation techniques (LoRA, QLoRA, P-Tuning) to specialize LLMs with limited resources.
- Computer Vision and Multimodality: Image, video, and audio processing using deep neural networks.
Academic-Practical Approach
We combine scientific research rigor with engineering pragmatics. You will learn to read arXiv papers and translate them into functional and optimized implementations.