Workshop: Machine Learning and Deep Learning

BloomTusks Technologies, we take pride in preparing both students and professionals with the cutting-edge skills needed to thrive in today’s tech-driven world. Continuing our commitment to practical and impactful learning, we recently organized a comprehensive workshop on Machine Learning, Deep Learning, and Generative AI.

The session was expertly delivered by Mr. Santhana Krishnan, who guided participants through key concepts, real-world use cases, and hands-on algorithmic demonstrations. The workshop served as a powerful introduction to the technologies shaping the future of AI and data science.

What is Machine Learning?

Machine Learning (ML) is a subset of Artificial Intelligence (AI) that enables systems to learn from data and improve over time without being explicitly programmed. It lies at the intersection of AI and data science, uncovering patterns and making predictions or decisions.

Examples:

• • Email spam detection

• • Movie recommendations

• • Fraud detection

Key Topics Covered in the ML

Types of Machine Learning:

• • Supervised Learning

• • Unsupervised Learning

• • Reinforcement Learning

ML Algorithms Overview:

• • Linear Regression

• • Logistic Regression

• • Decision Trees

• • K-Nearest Neighbors

• • K-Means Clustering

ML Pipeline:

• • Data Collection

• • Data Preprocessing

• • Model Training

• • Model Evaluation

• • Model Deployment

Model Evaluation Metrics:

Metric	Description
Confusion Matrix	Shows True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN)
Accuracy	Measures overall correctness
Precision / Recall / F1 Score	Useful for imbalanced datasets
ROC Curve	Visualizes classifier performance
MAE / MSE / RMSE	Error metrics for regression tasks

Hands-On with Real-Time Scenarios:

With the expert guidance of our speaker Mr. Santhana Krishnan, participants explored real-world industry scenarios through interactive case studies and practical simulations. The workshop was designed to go beyond theory offering hands-on exposure to how AI is applied across various domains:

• • In healthcare, learners understood how machine learning models can assist in disease prediction and accelerate drug discovery using patient datasets.

• • In the banking sector, they implemented basic fraud detection logic and learned how credit scoring models are developed.

• • For retail, participants explored how recommendation systems personalize customer experiences and how AI supports inventory forecasting.

• • In manufacturing, they examined predictive maintenance models and used deep learning (CNNs) for defect detection through visual inspection.

These sessions, clearly explained and guided by Mr. Santhana Krishnan, enabled participants to connect machine learning concepts with real-world outcomes, preparing them to apply their skills in practical environments.

What is Deep Learning?

Deep Learning is a subfield of ML that uses neural networks with multiple layers to model complex patterns in large datasets. It requires significant data and computational power.

Key Concepts Covered in Deep Learning

Applications of Deep Learning:

• • Facial Recognition: Security, auto-tagging on social media

• • Voice Assistants: Alexa, Siri

• • Autonomous Vehicles: Object detection, navigation

• • Language Translation: Google Translate, real-time interpretation

CNNs (Convolutional Neural Networks)

• • Used for: Image and video recognition

• • Components:
    
    • • Convolutional Layers: Extract features
    • • Pooling Layers: Down sampling
    • • Flattening: Converts data for dense layers

RNNs (Recurrent Neural Networks)

• • Used for: Sequential data (text, time series)

• • Advanced Models:
    

     

    • • LSTM (Long Short-Term Memory)
    • • GRU (Gated Recurrent Unit)
    • • Solve vanishing gradient problems and retain long-term memory

Neural Network Basics:

• • Perceptron: Basic computational unit

• • Network Structure: Input → Hidden → Output layers

• • Activation Functions: ReLU, Sigmoid, Tanh (introduce non-linearity)

What is Generative AI?

Generative AI refers to AI models that can create new content—including text, images, audio, video, and code—by learning patterns from existing data.

Examples:

• • ChatGPT for text generation

• • DALL·E for image synthesis

What is an LLM (Large Language Model)?

An LLM is a type of neural network trained on massive amounts of textual data using transformer architecture. These models understand and generate human-like language.

Examples: GPT, BERT, LLaMA

Architecture of LLMs:

• • Transformer: Employs self-attention for context understanding

• • Encoder-Decoder: Encoder processes input, decoder generates output

• • Training: Uses billions of parameters and diverse datasets

What is ChatGPT?

ChatGPT, developed by OpenAI, is a conversational AI based on the GPT architecture. It’s fine-tuned using Reinforcement Learning from Human Feedback (RLHF) for more natural and helpful responses.

How ChatGPT Works:

• • Tokenization: Breaks text into units (tokens)

• • Context Window: Remembers recent conversation for coherent replies

• • Response Generation: Predicts next tokens using trained probabilities

    Ethical Considerations:

• • Bias: May reflect training data bias

• • Misinformation: Can generate incorrect but plausible-sounding content

• • AI Safety: Requires human oversight and responsible usage

AI in Action: Platform Use Cases:

Platform	ML/AI Usage Area	Key Algorithms & Models
Facebook	Feed, Face Tagging, Ads	CNNs, DNNs, GBDTs, Transformers
WhatsApp	Spam & Security	Decision Trees, Anomaly Detection
Twitter	Feed, Toxicity, Bots	Transformers, Graph ML, Logistic Regression
Instagram	Explore, Moderation	CNNs, NLP models, Deep Ranking
YouTube	Recommendations, Ads	DNNs, Reinforcement Learning, BERT
TikTok	Video feed, Moderation	Deep Learning, RNNs, Reinforcement Learning
Snapchat	AR filters, Feed	CNNs, Computer Vision, Pose Estimation

Looking Ahead:

Workshops like these are just the beginning. At BloomTusks Technologies, we continue to build bridges between academic knowledge and industry needs by bringing real-world AI expertise into the classroom.

Stay tuned for our upcoming programs, and reach out if you’d like to join us on the path to mastering AI and Data Science!