Example: In a self-driving car simulation, reinforcement learning can be used to train the car to navigate roads by maximizing the reward (e.g., staying on the road, avoiding collisions), with deep Q-learning enabling it to learn from high-dimensional inputs like images from cameras.
How would you approach a situation where your model performs well on the training data but poorly on the test data?
Answer: This situation indicates overfitting. The approach to address it includes:
Cross-validation: To ensure that the model generalizes well to unseen data.
Regularization: Techniques like L1 or L2 regularization to penalize complex models and reduce overfitting.
Simplifying the model: Reducing the complexity by decreasing the number of features or layers in the model.
Gathering more data: To help the model learn a broader range of patterns.
Ensembling: Using methods like bagging or boosting to improve model robustness.
Example: If a neural network overfits on a small dataset, adding dropout layers during training or using a simpler model might help it generalize better to the test data.
What is transfer learning, and when would you use it?
Answer: Transfer learning is a technique where a model developed for a particular task is reused as the starting point for a model on a second task. It is particularly useful when the second task has limited data. Instead of training a model from scratch, the pre-trained model’s knowledge is transferred, and only the final layers are fine-tuned.
Example: In image classification, using a pre-trained model like VGG16 on ImageNet and fine-tuning it for a specific task like identifying specific animals can lead to good performance even with a small dataset.
To Summarize
Preparing for a machine learning interview requires a balance of theoretical knowledge and practical understanding. The questions covered in this article highlight key areas of machine learning, from foundational principles like the curse of dimensionality and loss functions to more complex topics like reinforcement learning and transfer learning. By familiarizing yourself with these questions and their corresponding answers, you’ll be better positioned to demonstrate your expertise and critical thinking skills during an interview.
How would you approach a situation where your model performs well on the training data but poorly on the test data?
Answer: This situation indicates overfitting. The approach to address it includes:
Cross-validation: To ensure that the model generalizes well to unseen data.
Regularization: Techniques like L1 or L2 regularization to penalize complex models and reduce overfitting.
Simplifying the model: Reducing the complexity by decreasing the number of features or layers in the model.
Gathering more data: To help the model learn a broader range of patterns.
Ensembling: Using methods like bagging or boosting to improve model robustness.
Example: If a neural network overfits on a small dataset, adding dropout layers during training or using a simpler model might help it generalize better to the test data.
What is transfer learning, and when would you use it?
Answer: Transfer learning is a technique where a model developed for a particular task is reused as the starting point for a model on a second task. It is particularly useful when the second task has limited data. Instead of training a model from scratch, the pre-trained model’s knowledge is transferred, and only the final layers are fine-tuned.
Example: In image classification, using a pre-trained model like VGG16 on ImageNet and fine-tuning it for a specific task like identifying specific animals can lead to good performance even with a small dataset.
To Summarize
Preparing for a machine learning interview requires a balance of theoretical knowledge and practical understanding. The questions covered in this article highlight key areas of machine learning, from foundational principles like the curse of dimensionality and loss functions to more complex topics like reinforcement learning and transfer learning. By familiarizing yourself with these questions and their corresponding answers, you’ll be better positioned to demonstrate your expertise and critical thinking skills during an interview.
DoorDash now uses RAG to solve complex issues for their Dashers accurately.
Here’s how --
The first step is to identify the core issue the Dasher is facing based on their messages, so that the system can retrieve articles on previous similar cases.
They break down complex tasks in the Dasher’s request, and use chain-of-thought prompting to generate an answer for it.
The most interesting part is their LLM Guardrail system:
1. First it applies a semantic similarity check between responses and knowledge base articles.
2. If the response fails, an LLM-powered evaluation checks for grounding, coherence, and compliance, so that it can prevent hallucinations and escalate problematic cases.
The knowledge base is continuously updated for completeness and accuracy, guided by the LLM quality assessments. For regression prevention they benchmark prompt changes before deployment as well.
Here’s how --
The first step is to identify the core issue the Dasher is facing based on their messages, so that the system can retrieve articles on previous similar cases.
They break down complex tasks in the Dasher’s request, and use chain-of-thought prompting to generate an answer for it.
The most interesting part is their LLM Guardrail system:
1. First it applies a semantic similarity check between responses and knowledge base articles.
2. If the response fails, an LLM-powered evaluation checks for grounding, coherence, and compliance, so that it can prevent hallucinations and escalate problematic cases.
The knowledge base is continuously updated for completeness and accuracy, guided by the LLM quality assessments. For regression prevention they benchmark prompt changes before deployment as well.
Suppose you're tasked with building a recommendation system for Instagram's Stories. How would you design the architecture?
This is a system-design question you get for Applied Data Science or Machine Learning Engineer rounds.
These rounds are typically interactive where you want to understand what specific problem the interviewer wants you to solve.
A system like stories needs to be highly scalable and handle-real time updates.
Keeping this in mind a few architecture aspect you could dive into are -
𝐃𝐚𝐭𝐚 𝐈𝐧𝐠𝐞𝐬𝐭𝐢𝐨𝐧 𝐋𝐚𝐲𝐞𝐫
Needs to handle streaming data from multiple sources:
Event Streams: User interactions (views, likes, skips, etc.) and creator activity.
Content Metadata: Story metadata like geotags, captions, etc.
Real-Time Features: Temporal features like time since Story creation.
User Features: Long-term (preferences, demographics) and short-term (session-based) behavioral data.
Kafka for streaming, Spark Streaming for processing, and a NoSQL database like DynamoDB for fast feature lookups - could be a good tech stack
𝐂𝐚𝐧𝐝𝐢𝐝𝐚𝐭𝐞 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧
Narrow down the pool of millions of Stories to hundreds of candidates.
👉 Compute embeddings for users and Stories using models like two-tower architectures - Deep Average Network (DAN) for user embeddings based on interaction history, Story embeddings derived from metadata and visual features (using models like CLIP or Vision Transformers).
👉 Use ANN (Approximate Nearest Neighbors) methods (e.g., ScaNN) to retrieve top-N candidates efficiently.
𝐑𝐚𝐧𝐤𝐢𝐧𝐠 𝐋𝐚𝐲𝐞𝐫
Score and rank the candidates for personalization.
👉 Gradient Boosted Decision Trees (e.g., LightGBM) or BERT4Rec to capture sequence-level dependencies in user interactions.
👉 Use bandit algorithms (e.g., Thompson Sampling or CMABs) to explore less popular Stories while exploiting known preferences.
𝐄𝐯𝐚𝐥𝐮𝐚𝐭𝐢𝐨𝐧
👉 Offline Metrics - NDCG (Normalized Discounted Cumulative Gain), and diversity scores.
👉 Online Metrics - View-through rate, engagement time, and session duration.
This is a system-design question you get for Applied Data Science or Machine Learning Engineer rounds.
These rounds are typically interactive where you want to understand what specific problem the interviewer wants you to solve.
A system like stories needs to be highly scalable and handle-real time updates.
Keeping this in mind a few architecture aspect you could dive into are -
𝐃𝐚𝐭𝐚 𝐈𝐧𝐠𝐞𝐬𝐭𝐢𝐨𝐧 𝐋𝐚𝐲𝐞𝐫
Needs to handle streaming data from multiple sources:
Event Streams: User interactions (views, likes, skips, etc.) and creator activity.
Content Metadata: Story metadata like geotags, captions, etc.
Real-Time Features: Temporal features like time since Story creation.
User Features: Long-term (preferences, demographics) and short-term (session-based) behavioral data.
Kafka for streaming, Spark Streaming for processing, and a NoSQL database like DynamoDB for fast feature lookups - could be a good tech stack
𝐂𝐚𝐧𝐝𝐢𝐝𝐚𝐭𝐞 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧
Narrow down the pool of millions of Stories to hundreds of candidates.
👉 Compute embeddings for users and Stories using models like two-tower architectures - Deep Average Network (DAN) for user embeddings based on interaction history, Story embeddings derived from metadata and visual features (using models like CLIP or Vision Transformers).
👉 Use ANN (Approximate Nearest Neighbors) methods (e.g., ScaNN) to retrieve top-N candidates efficiently.
𝐑𝐚𝐧𝐤𝐢𝐧𝐠 𝐋𝐚𝐲𝐞𝐫
Score and rank the candidates for personalization.
👉 Gradient Boosted Decision Trees (e.g., LightGBM) or BERT4Rec to capture sequence-level dependencies in user interactions.
👉 Use bandit algorithms (e.g., Thompson Sampling or CMABs) to explore less popular Stories while exploiting known preferences.
𝐄𝐯𝐚𝐥𝐮𝐚𝐭𝐢𝐨𝐧
👉 Offline Metrics - NDCG (Normalized Discounted Cumulative Gain), and diversity scores.
👉 Online Metrics - View-through rate, engagement time, and session duration.
Fundamentals of a 𝗩𝗲𝗰𝘁𝗼𝗿 𝗗𝗮𝘁𝗮𝗯𝗮𝘀𝗲.
With the rise of GenAI, Vector Databases skyrocketed in popularity. The truth is that a Vector Database is also useful for different kinds of AI Systems outside of a Large Language Model context.
When it comes to Machine Learning, we often deal with Vector Embeddings. Vector Databases were created to perform specifically well when working with them:
➡️ Storing.
➡️ Updating.
➡️ Retrieving.
When we talk about retrieval, we refer to retrieving set of vectors that are most similar to a query in a form of a vector that is embedded in the same Latent space. This retrieval procedure is called Approximate Nearest Neighbour (ANN) search.
A query here could be in a form of an object like an image for which we would like to find similar images. Or it could be a question for which we want to retrieve relevant context that could later be transformed into an answer via a LLM.
Let’s look into how one would interact with a Vector Database:
𝗪𝗿𝗶𝘁𝗶𝗻𝗴/𝗨𝗽𝗱𝗮𝘁𝗶𝗻𝗴 𝗗𝗮𝘁𝗮.
1. Choose a ML model to be used to generate Vector Embeddings.
2. Embed any type of information: text, images, audio, tabular. Choice of ML model used for embedding will depend on the type of data.
3. Get a Vector representation of your data by running it through the Embedding Model.
4. Store additional metadata together with the Vector Embedding. This data would later be used to pre-filter or post-filter ANN search results.
5. Vector DB indexes Vector Embedding and metadata separately. There are multiple methods that can be used for creating vector indexes, some of them: Random Projection, Product Quantization, Locality-sensitive Hashing.
6. Vector data is stored together with indexes for Vector Embeddings and metadata connected to the Embedded objects.
𝗥𝗲𝗮𝗱𝗶𝗻𝗴 𝗗𝗮𝘁𝗮.
7. A query to be executed against a Vector Database will usually consist of two parts:
➡️ Data that will be used for ANN search. e.g. an image for which you want to find similar ones.
➡️ Metadata query to exclude Vectors that hold specific qualities known beforehand. E.g. given that you are looking for similar images of apartments - exclude apartments in a specific location.
8. You execute Metadata Query against the metadata index. It could be done before or after the ANN search procedure.
9. You embed the data into the Latent space with the same model that was used for writing the data to the Vector DB.
10. ANN search procedure is applied and a set of Vector embeddings are retrieved. Popular similarity measures for ANN search include: Cosine Similarity, Euclidean Distance, Dot Product.
With the rise of GenAI, Vector Databases skyrocketed in popularity. The truth is that a Vector Database is also useful for different kinds of AI Systems outside of a Large Language Model context.
When it comes to Machine Learning, we often deal with Vector Embeddings. Vector Databases were created to perform specifically well when working with them:
➡️ Storing.
➡️ Updating.
➡️ Retrieving.
When we talk about retrieval, we refer to retrieving set of vectors that are most similar to a query in a form of a vector that is embedded in the same Latent space. This retrieval procedure is called Approximate Nearest Neighbour (ANN) search.
A query here could be in a form of an object like an image for which we would like to find similar images. Or it could be a question for which we want to retrieve relevant context that could later be transformed into an answer via a LLM.
Let’s look into how one would interact with a Vector Database:
𝗪𝗿𝗶𝘁𝗶𝗻𝗴/𝗨𝗽𝗱𝗮𝘁𝗶𝗻𝗴 𝗗𝗮𝘁𝗮.
1. Choose a ML model to be used to generate Vector Embeddings.
2. Embed any type of information: text, images, audio, tabular. Choice of ML model used for embedding will depend on the type of data.
3. Get a Vector representation of your data by running it through the Embedding Model.
4. Store additional metadata together with the Vector Embedding. This data would later be used to pre-filter or post-filter ANN search results.
5. Vector DB indexes Vector Embedding and metadata separately. There are multiple methods that can be used for creating vector indexes, some of them: Random Projection, Product Quantization, Locality-sensitive Hashing.
6. Vector data is stored together with indexes for Vector Embeddings and metadata connected to the Embedded objects.
𝗥𝗲𝗮𝗱𝗶𝗻𝗴 𝗗𝗮𝘁𝗮.
7. A query to be executed against a Vector Database will usually consist of two parts:
➡️ Data that will be used for ANN search. e.g. an image for which you want to find similar ones.
➡️ Metadata query to exclude Vectors that hold specific qualities known beforehand. E.g. given that you are looking for similar images of apartments - exclude apartments in a specific location.
8. You execute Metadata Query against the metadata index. It could be done before or after the ANN search procedure.
9. You embed the data into the Latent space with the same model that was used for writing the data to the Vector DB.
10. ANN search procedure is applied and a set of Vector embeddings are retrieved. Popular similarity measures for ANN search include: Cosine Similarity, Euclidean Distance, Dot Product.
👍1
Three different learning styles in machine learning algorithms:
1. Supervised Learning
Input data is called training data and has a known label or result such as spam/not-spam or a stock price at a time.
A model is prepared through a training process in which it is required to make predictions and is corrected when those predictions are wrong. The training process continues until the model achieves a desired level of accuracy on the training data.
Example problems are classification and regression.
Example algorithms include: Logistic Regression and the Back Propagation Neural Network.
2. Unsupervised Learning
Input data is not labeled and does not have a known result.
A model is prepared by deducing structures present in the input data. This may be to extract general rules. It may be through a mathematical process to systematically reduce redundancy, or it may be to organize data by similarity.
Example problems are clustering, dimensionality reduction and association rule learning.
Example algorithms include: the Apriori algorithm and K-Means.
3. Semi-Supervised Learning
Input data is a mixture of labeled and unlabelled examples.
There is a desired prediction problem but the model must learn the structures to organize the data as well as make predictions.
Example problems are classification and regression.
Example algorithms are extensions to other flexible methods that make assumptions about how to model the unlabeled data.
1. Supervised Learning
Input data is called training data and has a known label or result such as spam/not-spam or a stock price at a time.
A model is prepared through a training process in which it is required to make predictions and is corrected when those predictions are wrong. The training process continues until the model achieves a desired level of accuracy on the training data.
Example problems are classification and regression.
Example algorithms include: Logistic Regression and the Back Propagation Neural Network.
2. Unsupervised Learning
Input data is not labeled and does not have a known result.
A model is prepared by deducing structures present in the input data. This may be to extract general rules. It may be through a mathematical process to systematically reduce redundancy, or it may be to organize data by similarity.
Example problems are clustering, dimensionality reduction and association rule learning.
Example algorithms include: the Apriori algorithm and K-Means.
3. Semi-Supervised Learning
Input data is a mixture of labeled and unlabelled examples.
There is a desired prediction problem but the model must learn the structures to organize the data as well as make predictions.
Example problems are classification and regression.
Example algorithms are extensions to other flexible methods that make assumptions about how to model the unlabeled data.
👍2
🚦Top 10 Data Science Tools🚦
Here we will examine the top best Data Science tools that are utilized generally by data researchers and analysts. But prior to beginning let us discuss about what is Data Science.
🛰What is Data Science ?
Data science is a quickly developing field that includes the utilization of logical strategies, calculations, and frameworks to extract experiences and information from organized and unstructured data .
🗽Top Data Science Tools that are normally utilized :
1.) Jupyter Notebook : Jupyter Notebook is an open-source web application that permits clients to make and share archives that contain live code, conditions, representations, and narrative text .
2.) Keras : Keras is a famous open-source brain network library utilized in data science. It is known for its usability and adaptability.
Keras provides a range of tools and techniques for dealing with common data science problems, such as overfitting, underfitting, and regularization.
3.) PyTorch : PyTorch is one more famous open-source AI library utilized in information science. PyTorch also offers easy-to-use interfaces for various tasks such as data loading, model building, training, and deployment, making it accessible to beginners as well as experts in the field of machine learning.
4.) TensorFlow : TensorFlow allows data researchers to play out an extensive variety of AI errands, for example, image recognition , natural language processing , and deep learning.
5.) Spark : Spark allows data researchers to perform data processing tasks like data control, investigation, and machine learning , rapidly and effectively.
6.) Hadoop : Hadoop provides a distributed file system (HDFS) and a distributed processing framework (MapReduce) that permits data researchers to handle enormous datasets rapidly.
7.) Tableau : Tableau is a strong data representation tool that permits data researchers to make intuitive dashboards and perceptions. Tableau allows users to combine multiple charts.
8.) SQL : SQL (Structured Query Language) SQL permits data researchers to perform complex queries , join tables, and aggregate data, making it simple to extricate bits of knowledge from enormous datasets. It is a powerful tool for data management, especially for large datasets.
9.) Power BI : Power BI is a business examination tool that conveys experiences and permits clients to make intuitive representations and reports without any problem.
10.) Excel : Excel is a spreadsheet program that broadly utilized in data science. It is an amazing asset for information the board, examination, and visualization .Excel can be used to explore the data by creating pivot tables, histograms, scatterplots, and other types of visualizations.
Here we will examine the top best Data Science tools that are utilized generally by data researchers and analysts. But prior to beginning let us discuss about what is Data Science.
🛰What is Data Science ?
Data science is a quickly developing field that includes the utilization of logical strategies, calculations, and frameworks to extract experiences and information from organized and unstructured data .
🗽Top Data Science Tools that are normally utilized :
1.) Jupyter Notebook : Jupyter Notebook is an open-source web application that permits clients to make and share archives that contain live code, conditions, representations, and narrative text .
2.) Keras : Keras is a famous open-source brain network library utilized in data science. It is known for its usability and adaptability.
Keras provides a range of tools and techniques for dealing with common data science problems, such as overfitting, underfitting, and regularization.
3.) PyTorch : PyTorch is one more famous open-source AI library utilized in information science. PyTorch also offers easy-to-use interfaces for various tasks such as data loading, model building, training, and deployment, making it accessible to beginners as well as experts in the field of machine learning.
4.) TensorFlow : TensorFlow allows data researchers to play out an extensive variety of AI errands, for example, image recognition , natural language processing , and deep learning.
5.) Spark : Spark allows data researchers to perform data processing tasks like data control, investigation, and machine learning , rapidly and effectively.
6.) Hadoop : Hadoop provides a distributed file system (HDFS) and a distributed processing framework (MapReduce) that permits data researchers to handle enormous datasets rapidly.
7.) Tableau : Tableau is a strong data representation tool that permits data researchers to make intuitive dashboards and perceptions. Tableau allows users to combine multiple charts.
8.) SQL : SQL (Structured Query Language) SQL permits data researchers to perform complex queries , join tables, and aggregate data, making it simple to extricate bits of knowledge from enormous datasets. It is a powerful tool for data management, especially for large datasets.
9.) Power BI : Power BI is a business examination tool that conveys experiences and permits clients to make intuitive representations and reports without any problem.
10.) Excel : Excel is a spreadsheet program that broadly utilized in data science. It is an amazing asset for information the board, examination, and visualization .Excel can be used to explore the data by creating pivot tables, histograms, scatterplots, and other types of visualizations.
❤1
Walmart built an AI semantic search system processing millions of queries with 99% recall.
- When a user searches for a product, the query goes through a Siamese network with pre-trained tokenisers. This architecture allows the model to use the context of the input queries effectively.
- Different attributes are concatenated to the query title using a special token. This ensures that the model can distinguish among different product characteristics, like brand or colour, when processing a query.
- During training, the model employs a sampled softmax loss function, where both relevant and irrelevant products are considered for each query -- this helps improve the accuracy in distinguishing between different product matches.
- The architecture combines multiple embeddings for both queries and products. This lets the system capture the varying meanings of common queries, improving the model's flexibility and interpretation.
- When a user searches for a product, the query goes through a Siamese network with pre-trained tokenisers. This architecture allows the model to use the context of the input queries effectively.
- Different attributes are concatenated to the query title using a special token. This ensures that the model can distinguish among different product characteristics, like brand or colour, when processing a query.
- During training, the model employs a sampled softmax loss function, where both relevant and irrelevant products are considered for each query -- this helps improve the accuracy in distinguishing between different product matches.
- The architecture combines multiple embeddings for both queries and products. This lets the system capture the varying meanings of common queries, improving the model's flexibility and interpretation.
Data Science Interview Question
Q. What is the difference between bagging and boosting in decision trees?
Answer:
Bagging or bootstrap aggregation, is the process of randomly sampling with replacement multiple times from your original dataset and fitting a decision tree to each of the datasets. Then, you average the predictions made across multiple datasets for regression, or take the majority vote for classification. This averaging approach reduces the variance of the ensemble, and performs better than a single decision tree which fits the data hard and is likely to overfit.
Boosting, on the other hand, is a sequential learning approach. Given the original dataset, the boosting approach does not attempt to fit the data hard, but learns slowly. In boosting, given the current model, the algorithm fits a small, shrunken tree to the residuals of the model. Then, it adds this shrunken tree to the original tree to update the residuals. It continues this process as more small, shrunken trees are fit to the residuals of the model. By focusing on improving residual error, and using this stepwise, sequential approach, the function improves in areas that it usually does not perform well in.
Q. What is the difference between bagging and boosting in decision trees?
Answer:
Bagging or bootstrap aggregation, is the process of randomly sampling with replacement multiple times from your original dataset and fitting a decision tree to each of the datasets. Then, you average the predictions made across multiple datasets for regression, or take the majority vote for classification. This averaging approach reduces the variance of the ensemble, and performs better than a single decision tree which fits the data hard and is likely to overfit.
Boosting, on the other hand, is a sequential learning approach. Given the original dataset, the boosting approach does not attempt to fit the data hard, but learns slowly. In boosting, given the current model, the algorithm fits a small, shrunken tree to the residuals of the model. Then, it adds this shrunken tree to the original tree to update the residuals. It continues this process as more small, shrunken trees are fit to the residuals of the model. By focusing on improving residual error, and using this stepwise, sequential approach, the function improves in areas that it usually does not perform well in.
Walk away from people and situations where you do not get respect. When someone does not value you, just ignore them and move on. The more we walk away from such situations, the more we open ourselves up to future possibilities in life.
Machine Learning And AI
https://youtu.be/8r_yY23mpV0
Hi guys a lot of you have not subscribed my channel yet. If you're reading this message then don't forget to subscribe my channel and comment your views. At least half of you go and subscribe my channel.
Thank you in advance
Thank you in advance
https://youtu.be/HCZPOuxYi3I
Hi guys a lot of you have not subscribed my channel yet. If you're reading this message then don't forget to subscribe my channel and comment your views. At least half of you go and subscribe my channel.
Thank you in advance
Hi guys a lot of you have not subscribed my channel yet. If you're reading this message then don't forget to subscribe my channel and comment your views. At least half of you go and subscribe my channel.
Thank you in advance
YouTube
SQL Query to Calculate Average Processing Time Per Machine 🚀| Leetcode 1661
SQL Query to Calculate Average Processing Time Per Machine 🚀
**Description:**
In this video, we solve a **SQL interview question** on calculating the **average processing time per machine**. We use **GROUP BY, CASE WHEN, and aggregation functions** to…
**Description:**
In this video, we solve a **SQL interview question** on calculating the **average processing time per machine**. We use **GROUP BY, CASE WHEN, and aggregation functions** to…