DeepLearning.AI Specialization — Summary of Certifications

Neural Networks and Deep Learning

Instructor(s): Andrew Ng, Kian Katanforoosh & Younes Bensouda Mourri
Completed by: Anish Shrestha
Completion Date: January 11, 2023
Duration: ~5 weeks (5 hours/week)

Summary:
Introduced the foundations of deep learning, neural network architectures, forward/backpropagation, and key concepts like gradient descent and activation functions. Built basic neural network models and understood how deep learning is applied in real-world tasks.

---

Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

Instructor(s): Andrew Ng, Kian Katanforoosh & Younes Bensouda Mourri
Completed by: Anish Shrestha
Completion Date: February 1, 2024
Duration: ~5 weeks (5 hours/week)

Summary:
Focused on advanced model-building techniques including hyperparameter tuning, optimization algorithms (Adam, RMSProp, Momentum), and regularization methods like dropout. Learned practical techniques to improve neural network performance.

---

Structuring Machine Learning Projects

Instructor(s): Andrew Ng, Younes Bensouda Mourri & Kian Katanforoosh
Completed by: Anish Shrestha
Completion Date: February 15, 2024
Duration: ~4 weeks (5 hours/week)

Summary:
Covered best practices for designing ML workflows, handling training data challenges, error analysis, and prioritizing tasks in ML projects. Developed skills for troubleshooting learning problems and improving model deployment efficiency.

---

Convolutional Neural Networks

Instructor(s): Andrew Ng, Kian Katanforoosh & Younes Bensouda Mourri
Completed by: Anish Shrestha
Completion Date: March 7, 2024
Duration: ~5 weeks (5 hours/week)

Summary:
Learned the architecture and theory behind CNNs and their applications in computer vision. Implemented convolutional models, transfer learning, detection, and segmentation tasks for image-based AI systems.

---

Sequence Models

Instructor(s): Andrew Ng, Kian Katanforoosh & Younes Bensouda Mourri
Completed by: Anish Shrestha
Completion Date: April 21, 2024
Duration: ~5 weeks (5 hours/week)

Summary:
Explored recurrent neural networks (RNNs), LSTMs, GRUs, and sequence-to-sequence models. Applied these methods to natural language processing, speech recognition, and time-series tasks. Also covered attention mechanisms and practical sequence model design.

---