Advanced Certification in Deep Learning – IISc Bangalore

This module covers essential mathematical foundations and visualization techniques needed for machine learning, including linear algebra concepts, probability and statistics fundamentals, numerical optimization methods for ML, and comprehensive data visualization approaches.

This module covers essential mathematical foundations and visualization techniques needed for machine learning, including linear algebra concepts, probability and statistics fundamentals, numerical optimization methods for ML, and comprehensive data visualization approaches.

This module provides comprehensive coverage of core deep learning architectures and techniques, including supervised and unsupervised learning paradigms, neural network fundamentals such as CNNs, RNNs, and LSTMs, backpropagation mechanisms, deep network design principles, and advanced regularization methods including dropout and batch normalization.

This module provides comprehensive coverage of core deep learning architectures and techniques, including supervised and unsupervised learning paradigms, neural network fundamentals such as CNNs, RNNs, and LSTMs, backpropagation mechanisms, deep network design principles, and advanced regularization methods including dropout and batch normalization.

This module explores deep learning techniques specifically designed for speech and audio processing applications, including audio representation methods for deep learning, speech recognition systems, and end-to-end deep network architectures optimized for speech-related tasks.

This module covers deep learning applications in computer vision, including object detection models, popular CNN architectures, transfer learning techniques, autoencoders, advanced tasks like object detection and image segmentation, recurrent architectures for image captioning and video analysis, and introduction to sequential decision making under uncertainty.

This module introduces reinforcement learning concepts and their integration with deep neural networks, covering practical implementation of RL algorithms using deep learning, value functions for decision making, and problem formulations for both finite and infinite horizon scenarios.

This module focuses on implementing deep learning models on resource-constrained IoT and edge computing devices, covering overview of various ML hardware platforms for IoT/Edge applications, energy efficiency considerations, optimization techniques for edge deployment, and ML model adaptations for edge computing environments.

This module covers advanced representation learning techniques including deep generative models across two parts for synthetic data generation, and semi-supervised and self-supervised learning approaches across two parts that reduce reliance on labeled data and enable learning from partially labeled or unlabeled datasets.

This module provides hands-on practical experience through curated capstone projects covering diverse deep learning applications including COVID-19 detection using acoustic sounds, NLP models for Math Word Problems, BERT models for Sequence Labeling, Prompt Learning for Vision Language Models, speech recognition and synthesis, voice conversion, reinforcement learning for mobile edge computing, anomaly detection in Industrial IoT, federated recommender systems, GAN controls, adversarial samples, depth estimation, image captioning, zero-shot segmentation and classification, knowledge distillation, time series forecasting via LLMs, lesion segmentation, legal judgement prediction, and optic disk-cup segmentation.