Schedule - AIR.25
| Days / Hours | 1030-1300 | 1300-1430 | 1430-1700 |
| MONDAY | Keynote Lecture and Inauguration | L U N C H | Fundamental of Neural Network (NN) |
| TUESDAY | Fundamental of Deep Learning (DL) | Python Programming (Part-I) | |
| WEDNESDAY | Fundamental of CNN and RNN | Python Programming (Part-II) | |
| THURSDAY | Machine Learning Libraries (Part-I) | Machine Learning Libraries (Part-II) | |
| FRIDAY | Machine Learning Libraries (Part-III) | Data Preprocessing Techniques | |
| SATURDAY | Machine Learning Model (Part-I) | Machine Learning Model (Part-II) | |
| SUNDAY | Week Off | Week Off |
| Days / Hours | 1030-1300 | 1300-1430 | 1430-1700 |
| MONDAY | Natural Language Processing (NLP): Tokenization, text cleaning, embeddings | L U N C H | Transformer Models: Attention mechanism, encoder-decoder architecture |
| TUESDAY | Prompt Engineering: Designing effective prompts for LLMs | Semantic Search: Dense retrieval using embeddings | |
| WEDNESDAY | Traditional Search: TF-IDF, BM25 (can be combined with RAG) | Hybrid Retrieval: Mixing keyword and vector-based search for accuracy | |
| THURSDAY | Vector Databases and Indexing | Frontend and Interfaces | |
| FRIDAY | Agentic RAG: Using tools, memory, or multi-tool agents to interact with sources | Local chatbot development like ChatGPT | |
| SATURDAY | Doubt Clearing Session (On Demand) | Week Off | |
| SUNDAY | Week Off | Week Off |
| Days / Hours | 1030-1300 | 1300-1430 | 1430-1700 |
| MONDAY | Generative Adversial Network | L U N C H | Generative Adversial Network |
| TUESDAY | Embedded System for Image Processing | Reinforcement Learning | |
| WEDNESDAY | Reinforcement Learning | Reinforcement Learning | |
| THURSDAY | Autonomous Navigation, SLAM and Its Classification | RTABmap to Mapping | |
| FRIDAY | ORB SLAM3 | Graph Based SLAM | |
| SATURDAY | Doubt Clearing Session (On Demand) | Week Off | |
| SUNDAY | Week Off | Week Off |
| Days / Hours | 1030-1300 | 1300-1430 | 1430-1700 |
| MONDAY | Point Cloud Processing and Reconstruction | L U N C H | Point Cloud Stitchng and Environment Creation |
| TUESDAY | Autonomous Mobile Robots, Simultanious Localization and Mapping | Simulations and ROS | |
| WEDNESDAY | Motion Planning, Learning-based Robotics | Control and Navigation with ROS | |
| THURSDAY | Robotics Hardware (Part-I) | Robotics Hardware (Part-II) | |
| FRIDAY | Robotics Hardware (Part-III) | Certificate Distribution (30 Days Training) | |
| SATURDAY | Week Off | Week Off | |
| SUNDAY | Week Off | Week Off |
| Days / Hours | 1700-1900 | Remark(s) |
| MONDAY | Define the Problem and Goal | The first step is to identify a real-world problem you want to address with ML. This could be anything from spam email detection to stock price prediction. Clearly define what your project aims to achieve. |
| TUESDAY | Data Collection | Machine learning thrives on data. You'll need to collect relevant data for your project. This could involve scraping data from the web, using public datasets, or generating your own data. Ensure the data is high-quality and aligns with your project's goal. |
| WEDNESDAY | Data Preprocessing | Raw data often needs cleaning and preparation before feeding it to an ML model. This might involve handling missing values, removing outliers, converting data types, and formatting the data for your chosen algorithms. |
| THURSDAY | Exploratory Data Analysis (EDA) | Get familiar with your data! Use Python libraries like pandas and matplotlib to explore the data's distribution, identify patterns, and uncover relationships between features. This helps you understand your data better and choose suitable ML techniques. |
| FRIDAY | Model Selection | There are various ML models available, each suited for specific tasks. Common choices include linear regression, decision trees, and neural networks. Consider the problem you're tackling and the characteristics of your data when selecting a model. |
| SATURDAY | Week Off | Not Applicable |
| SUNDAY | Week Off | Not Applicable |
| Days / Hours | 1700-1900 | Remark(s) |
| MONDAY | Model Training | Split your data into training and testing sets. The training set is used to train your model, while the testing set evaluates its performance on unseen data. Train your model using the chosen algorithm and the training data. |
| TUESDAY | Model Evaluation | Once trained, assess your model's performance on the testing set. Metrics like accuracy, precision, and recall help gauge how well your model generalizes to unseen data. |
| WEDNESDAY | Model Improvement | Evaluation might reveal areas for improvement. You can try tuning hyperparameters, adjusting the model architecture, or even exploring different algorithms altogether. |
| THURSDAY | Deployment | If your model performs well, consider deploying it for real-world use. This might involve integrating it into a web application, creating a standalone service, or even deploying it on mobile devices. |
| FRIDAY | Certificate Distribution (Online) | Hard Copy of the Certificate will be delivered by the post on your given corresponding address. |
| SATURDAY | Week Off | Not Applicable |
| SUNDAY | Week Off | Not Applicable |