Machine Learning Engineer Concept Interview Questions

Megha V. - "Gradient Descent is an optimisation strategy used in several supervised learning models. It is the technique for finding the optimum solution of an objective function. Typically, for a linear regression use case, it is used to find the weights and bias that produce the lowest loss. It involves computing the partial derivative of the objective function with respect to the weight and bias vectors. To find the optima of the function, the derivative is equated to 0, and iteratively the weight and b"See full answer

Louie Z. - "DNNs can learn hierarchical features, with each layer learning progressively more abstract features, and generalizes better. SNNs are better for simplier problems involving smaller datasets and if low latency is required."See full answer

Anonymous Aardvark - "in simple words, linear regression helps in predicting the value whereas logistics regression helps in predicting the binary classification. But lets talk through some example Linear regression model: E-commerce website pricing recommendation engine is built on linear regression model where we do have some variables such as competitor price, internal economics and consumer demand etc when we put this in a supervised learning model, it helps in predicting prices Logistics regression model"See full answer

Will I. - "Is it bad to get the answer a different way? Will they mark that as not knowing Bayes Theorem or just correct as it is an easier way to get the answer? The way I went is to look at what happens when the factory makes 100 light bulbs. Machine A makes 60 of which 3 are faulty, Machine B makes 40 of which 1.2 are faulty. Therefore the pool of faulty lightbulbs is 3/4.2 = 5/7 from machine A and 1.2/4.2 = 3/7 from Machine B."See full answer

Praveen D. - "Picking the value of 'k' can be little tricky. One simple approach would be to start with elbow method. Try different 'k' values and plot them on a graph like an experiment. When the graph starts to look like an elbow, that's a good 'k' to pick! Also keep in mind you may not always have clearly clustered data. The elbow may not be clear and sharp as you expect. Hope this helps."See full answer

Jyoti V. - "Over-fitting of a model occurs when model fails to generalize to any new data and has high variance withing training data whereas in under fitting model isn't able to uncover the underlying pattern in the training data and high bias. Tree based model like decision tree and random forest are likely to overfit whereas linear models like linear regression and logistic regression tends to under fit. There are many reasons why a Random forest can overfits easily 1. Model has grown to its full depth a"See full answer

Constantin P. - "Reinforcement Learning is a type of machine learning where an agent learns to make decisions by trying out different actions and receiving rewards or penalties in return. The goal is to learn, over time, which actions yield the highest rewards. There are three core components in RL: The agent — the learner or decision-maker (e.g., an algorithm or robot), The environment — everything the agent interacts with, Actions and rewards — the agent takes actions, and the environmen"See full answer

Hari priya K. - "Hey Grandma, you've had a lot of experience with infants, haven't you? When they were babies, you taught them how to chew in their first six months. This initial phase is like giving them data. Once they learned how to chew, they could handle any food you gave them. Next, you refined their learning by teaching them that they should only chew on food. This is like refining the data so they understand what is relevant. Then, a few months later, they started crawling and walking, learning by observ"See full answer

Darpan D. - "Clarifying When we say cloud gaming, we refer to a video gaming experience using cloud computing, right? Assumption: Yes. Understanding of cloud computing first. I'll use some analogies: Imagine you are looking to do heavy computing but don't have a powerful CPU and GPU. CPU and GPU are like your big calculators. You can buy a powerful CPU and GPU, but problems: It costs a lot to buy. It costs a lot to run. You don't need it 24-7. You are not a un"See full answer

Surbhi G. - "Deep Learning is a part of Artificial Intelligence, it's like teaching the machine to think and make decisions on its own. It's like how we teach a child the concept of an apple - it's round, red, has a stem on top. We show them multiple pictures of apples and then they understand and can recognize an apple in future. Similarly, we feed lots of data to the machine, and slowly, it starts learning from that data, and can then make relevant predictions or decisions based on what it has learnt. A co"See full answer

Praveen D. - "Grandma! You know how we can look at a picture and know what's in it—like seeing a cat or a dog? Computers can learn to do that too! It's just they use special tricks and math to see and understand pictures or videos. It helps them figure out what's in the pictures, almost like how we do! Almost like giving it eyes to see the world in its own way!"See full answer

Vishwanath K. - "DevOps Engineer Interview Questions for 3+ yrs experience candidate"See full answer

Sanmitra I. - "BERT is a bidirectional encoder representation transformer. It takes a sequence of tokens and produces vector embeddings for each token. The BERT Model was trained on the task of next sentence prediction task and masked language modeling. The key difference between word2vec and BERT is that, word2vec produces semantic embeddings for each word, where as BERT produces contextual word embeddings based on the relationships between surrounding words. Example - Vector Embedding of lets say word "i"See full answer

Shibin P. - "A typical computer vision pipeline consists of several key stages that process and analyze visual data to extract meaningful information. Here’s a general outline of the steps involved: Image Acquisition:Capturing images or videos using cameras or other imaging devices. Preprocessing steps such as resizing, cropping, and converting color spaces. Image Preprocessing:Noise reduction (e.g., using filters like Gaussian blur). Image normalization to standardize pixel values. Contrast e"See full answer