| What is underfitting? | | Basic Concepts |
| F1 score | | Metrics |
| Why is L1 regularization supposed to lead to sparsity than L2? | | Regularization |
| What is backpropagation? | | NN |
| Why MSE doesn’t work with logistic regression? | | Metrics |
| Why AUC is desirable? | | Metrics |
| How to avoid overfitting? | | Basic Concepts |
| L1 VS L2 regularization | | Regularization |
| Generative vs Discriminative | | Basic Concepts |
| Lasso and Ridge | | Regularization |
| Hinge loss - SVM | | Loss |
| Why is bias used in NN? | | NN |
| Entropy / Cross-Entropy / Relative-entropy loss | | Loss |
| Cross Validation / Stratified cross-validation | | Basic Concepts |
| Confusion matrix | | Metrics |
| Why does regularization use L1, L2 not others? | | Regularization |
| Why can regularization reduce overfitting? | | Regularization |
| Compare two classifiers | | System design |
| What is overfitting? | | Basic Concepts |
| Precision and recall | | Metrics |
| SGD | | NN |
| Initializing weight with 0 for NN? | | NN |
| Sigmoid function | | Activation Function |
| Adam optimizer | | Training |
| Learning rate | | NN |
| tanh | | Activation Function |
| Grid search and random search | | Metrics |
| Why is non linear activation function needed? | | Activation Function |
| Relation between logistic regression and neural network? | | NN |
| Exploding gradient | | NN |
| Does NN fit data better than logistic regression? | | System design |
| Vanishing gradient | | NN |
| RMSprop | | NN |
| ReLU | | Activation Function |
| linear regression vs logistic regression | | Basic Concepts |
| Fourier transform | | Basic Concepts |
| 1x1 convolutional filter | | CNN |
| Bias metric | | Basic Concepts |
| Naive Bayes | | Basic Concepts |
| supervised, unsupervised, semi-supervised, weakly-supervised, Self-supervised and unsupervised learning reinforcement learning | | Basic Concepts |
| CNNs in segmentation | | CNN |
| data visualization libraries | | CV |
| model accuracy or model performance | | Basic Concepts |
| CV Framework description in Chinese | | CV |
| What is attention, why attention | | NN |
| equivariant translation， invariant translation | | CNN |
| Bayes’ Theorem How to interpret Bayes rule | | Basic Concepts |
| Residual Networks skip connections | | CNN |
| Parameter calculation | | CNN |
| Convolution / Kernel / Feature map dimension | | CNN |
| Small convolutional kernels such as 3x3 rather than a few large ones | | CNN |
| max-pooling CNN | | CNN |
| Gradient Descent | | Basic Concepts |
| Compare CNN and Transformer | | NN |
| Transformer | | NN |
| Data Drift Detection | Importance of Data Drift Detection | | Data prepare |
| one important feature missing from a trained model, what can we do? | | Data prepare |
| why convolution layer for images than FC | | CNN |
| calculate convolution layer output size | | CNN |
| RNN | | CNN |
| Deep learning | | NN |
| Autoencoder | | CNN |
| GAN | | CNN |
| LSTM | | CNN |
| Turing test | | Data prepare |
| Why do we need a validation set and test set | | Data prepare |
| data normalization | | Data prepare |
| How to deal with imbalanced dataset | | Data prepare |
| data augmentation | | Data prepare |
| PCA - Principal Components Analysis | | Basic Concepts |
| Dropouts | | Basic Concepts |
| Reinforcement | | Basic Concepts |
| Partition | | Basic Concepts |
| Attribute | | NN |
| Accuracy | | Basic Concepts |
| Logistic Regression | | Basic Concepts |
| Batch norm, mini-batch, layer norm | | NN |
| Softmax Activation Function | | Activation Function |
| ROC AUC | | Metrics |
| word2vec vs. doc2vec | | Basic Concepts |
| covariance matrix | | Metrics |
| Type I error / Type II error loss | | Loss |
| What is cost function loss function | | Loss |
| Receptive Field | | CNN |
| confidence interval | | Basic Concepts |
| Why do I have to convert "uint8" into "float32" | | Data prepare |
| Log loss | | Loss |
| Loss becomes Inf or NaN - reason | | Loss |
| F-beta Score | | Metrics |
| EM algorithm | | Basic Concepts |
| how you would create a 3D model of an object from imagery and depth sensor measurements taken at all angles around the object | | CV |
| How does CBIR work | | CV |
| Handling Outliers in data | | Data prepare |
| CNNs translation invariant | | CNN |
| RMSE | | Metrics |
| How do you prepare a data? | | Data prepare |
| Bias-Variance Tradeoff | | Regularization |
| Write Conv2D and Active function from scratch | | ML Coding |
| Write Max Pooling from scratch | | ML Coding |
| Ran‍‌‌‌‌‍‌‌‌‌‌‍‍‌‍‌dom Forest vs. Gradient Boosted Forest (Decision Tree) | | Basic Concepts |
| Maximum Likelihood Estimation (MLE) - when find the mean and standard diviation of a group of models | | Basic Concepts |
| K-means | | Basic Concepts |
| Compare RNN,LSTM with transformer | | NN |
| L1 loss vs. L2 loss | | Loss |
| Ensemble boosting bagging | | Basic Concepts |
| difference between t-SNE and UMAP for dimensionality reduction | | Basic Concepts |
| non maximal suppression | | CV |
| Label Encoding vs. One Hot Encoding | | Basic Concepts |
| Instance-Based vs Model-Based Learning | | Basic Concepts |
| t-SNE sklearn | | Basic Concepts |
| UMAP | | Basic Concepts |
| difference between LDA and PCA for dimensionality reduction | | Basic Concepts |
| PCA, t-SNE, and UMAP | | Basic Concepts |
| probability vs maximum likelihood | | Basic Concepts |
| KNN different from k-means clustering | | NN |
| Create a function to compute an integral image, and create another function to get area sums from the integral image | | CV |
| Epoch vs. Batch vs. Iteration | | Training |
| Batch Gradient Descent and Stochastic Gradient Descent | | NN |
| Momentum | | NN |
| Connected Component Labeling | | CV |
| Pytorch CNN | | ML Coding |
| Write plot image using Matplotlib pyplot from scratch | | ML Coding |
| Write K-means from scratch | | ML Coding |
| Write Gradient Descent / Active Function from scratch | | ML Coding |
| PyTorch Cheatsheet | | ML Coding |
| Write KNN from scratch | | ML Coding |
| Python Library | | ML Coding |
| Feature Crosses | | Basic Concepts |
| SOTA | | SOTA |
| Feature Hashing | | Basic Concepts |
| One-hot encoding | | Basic Concepts |
| Embedding | | Basic Concepts |
| Numeric Features | | Basic Concepts |
| How do we measure similarity? | | Metrics |
| Mean encoding | | Basic Concepts |
| GELU | | Activation Function |
| Loss Functions | | Loss |
| Huber loss | | Loss |
| Data Generation Strategy | | Training |
| Handle Imbalance Class Distribution | | Training |
| Data Partitioning | | Training |
| Mean square error and mean absolute error | | Loss |
| Common Resampling Use Cases | | Training |
| A/B Testing in Ads | | System design |
| How LinkedIn Generates Data for Course Recommendation | | System design |
| Conditional Random Fields (CRFs) | | Basic Concepts |
| Random Number Generator | | ML Coding |
| Split Train test valid | | ML Coding |
| Quantile Loss | | Loss |
| Active function | | ML Coding |
| parallel implementation | | ML Coding |
| Pytorch Skip connection | | ML Coding |
| Forcast Metrics | | Metrics |
| Normalized Cross Entropy | | Metrics |
| Focal loss | | Loss |
| Write a Perceptron Classifier for Binary Classification | | ML Coding |
| Write a Simple Image Classifier (Using package) | | ML Coding |
| Write a Naive Bayes Classifier from scratch | | ML Coding |
| IOU | | Metrics |
| KNN | | Basic Concepts |
| Activation functions | | Activation Function |
| DALL·E | | NN |
| Vision Transformer | | NN |
| MultiModal (Text-Image) | | NN |
| BERT | | NN |
| What do you understand by transfer learning? Name a few commonly used transfer learning models. | | Basic Concepts |
| 01 ML system design introduction - bytebytego | | System design |
| Write FC Layer Neural Network from scratch | | ML Coding |
| Retrieval Augmented Generation | | GENAI |
| AlBEF - BLIP - BLIP2 | | GENAI |
| MultiModal | | GENAI |
| PCA | | ML Coding |
| SVM (Support Vector Machine) | | ML Coding |
| Logistic Regression | | ML Coding |