😔 It didn't work out...

Facing the upcoming graduation in February 2024, I turned down the master's proposal from the unit's graduate school, and I've been rejected by every graduate school and company I applied to, leaving me lost.

🎓 Getting a job directly in ML with my unrefined knowledge and naive projects from undergrad seemed insufficient. Still, I have no regrets. The effort I put in led to meaningful project experiences, and the attempt to grasp various mathematical expressions will undoubtedly serve as a stepping stone for my growth.

A New Challenge

For the past 9 months, preparing for employment and grad school applications prevented me from studying what I truly wanted. I've applied to various companies and firms without regret, so now I want to dive back into the studies and projects I desire.

Starting by posting on Velog, I aim to study and consolidate knowledge on deep learning, FairAI, and XAI. With experiences from undergrad and two years as a research assistant, my goal is to organize the knowledge I naively possess and delve deeper.

🚀 First, I obtained keywords for study through ChatGPT. Now, I will study and organize the following keywords one by one.

100 Deep Learning Keywords / 50 Fair AI & XAI Keywords

I will add the organized content with LINKs soon.

Table 1: Deep Learning Fundamentals

No.	Topic	Content	LINK
1	Neuron	Understanding the operation up to the activation function
2	Activation Function	Comparisons and purposes of various functions
3	Feedforward Neural Network	Understanding the structure and operation
4	Backpropagation	Core principles of the error backpropagation algorithm
5	Gradient Descent	Principles and types of gradient descent
6	Loss Function	Understanding various loss functions and their purposes
7	Overfitting	Methods to prevent overfitting
8	Underfitting	Understanding methods to prevent underfitting
9	Epoch	Grasping the concept and importance of epochs
10	Batch Size	Role and criteria for selecting batch size
11	Learning Rate	Impact and adjustment methods of the learning rate
12	Weight Initialization	Importance and methods of weight initialization
13	Vanishing Gradient Problem	Understanding the problem of vanishing gradients
14	Exploding Gradient Problem	Understanding the problem of exploding gradients
15	Convolutional Neural Network (CNN)	Structure specialized in image processing
16	Pooling	Role and various types of pooling
17	Transfer Learning	Concept and application of transfer learning
18	Recurrent Neural Network (RNN)	Basic structure of recurrent neural networks
19	Long Short-Term Memory (LSTM)	LSTM for sequence learning
20	Gated Recurrent Unit (GRU)	Simple structure of the GRU, a type of RNN
21	Autoencoder	Basic principles of autoencoders
22	Generative Adversarial Network (GAN)	Understanding the generative model GAN
23	Word Embeddings	Concept of word embeddings and Word2Vec, GloVe
24	Attention Mechanism	Role and structure of attention mechanisms
25	Transformer	Transformer based on attention mechanisms
26	Reinforcement Learning	Basic concepts of reinforcement learning
27	Q-Learning	Operation and application of the Q-learning algorithm
28	Policy Gradient	Concept of policy gradients and its application
29	Actor-Critic	Structure and pros/cons of the actor-critic model
30	Batch Normalization	Role and advantages of batch normalization
31	Dropout	Purpose and effects of dropout
32	Hyperparameter Tuning	Methods for tuning hyperparameters
33	Ensemble Learning	Principles and types of ensemble learning
34	Bias-Variance Tradeoff	Bias-variance tradeoff
35	Implicit & Explicit	Understanding meaning in Mathmatics & ML	LINK
36	ROC Curve	Meaning and interpretation of the ROC curve	LINK
37	Precision and Recall	Difference and importance of precision and recall	LINK
38	F1 Score	Meaning and usage of the F1 score	LINK
39	Confusion Matrix	Composition and interpretation of confusion matrix
40	One-Hot Encoding	Purpose and application of one-hot encoding
41	Normalization and Standardization	Difference and usage of normalization and standardization
42	Data Augmentation	Concept and application of data augmentation
43	Imbalanced Data	Methods for handling imbalanced data
44	K-Nearest Neighbors (KNN)	Operating principles of the KNN algorithm
45	Decision Trees	Structure and learning principles of decision trees
46	Random Forest	Concept and advantages of random forests
47	Support Vector Machine (SVM)	Operating principles and applications of SVM
48	Principal Component Analysis (PCA)	Purpose and principles of PCA
49	Cross-Entropy Loss	Meaning and usage of cross-entropy loss
50	Mean Squared Error (MSE)	Meaning and usage of mean squared error
51	Activation Functions	Sigmoid, Tanh, and ReLU in-depth analysis
52	Normalization Techniques	Batch Normalization vs. Layer Normalization
53	Weight Regularization	L1 and L2 regularization for preventing overfitting
54	Optimizers	SGD, Adam, RMSprop - comparing optimization algorithms
55	Learning Rate Schedulers	Strategies for adapting learning rates during training
56	Data Preprocessing	Techniques for cleaning and preparing input data
57	Model Evaluation Metrics	Beyond accuracy - precision, recall, F1 score
58	Neural Architecture Search	Automated methods for finding optimal network structures
59	Explainable AI (XAI)	Advancements in making AI models more interpretable
60	Hyperparameter Optimization	Advanced techniques for tuning model parameters
61	Generative Models	Beyond GANs - Variational Autoencoders (VAEs)
62	Self-Supervised Learning	Training models without explicit labels
63	Adversarial Attacks	Methods for testing and improving model robustness
64	Federated Learning	Collaborative model training without centralized data
65	Quantum Machine Learning	The intersection of quantum computing and ML
66	Explainability in NLP	Interpretable models in Natural Language Processing
67	Graph Neural Networks	Extending neural networks to graph-structured data
68	Meta-Learning	Models that can learn how to learn
69	Capsule Networks	A novel approach to handling hierarchical relationships
70	Transfer Learning Strategies	Leveraging pre-trained models effectively
71	Attention Mechanism Variants	Multi-head attention, scaled dot-product attention
72	Memory-Augmented Networks	Neural networks with external memory structures
73	Deep Reinforcement Learning	Merging deep learning and reinforcement learning
74	Imitation Learning	Learning from observing demonstrations
75	Explainability in Image Classification	Interpreting CNNs for image classification tasks
76	Few-Shot Learning	Training models with very limited labeled examples
77	Energy-Based Models	A different paradigm for defining probabilistic models
78	AI in Healthcare	Applications and challenges in medical settings
79	AI in Finance	Utilizing AI for financial modeling and analysis
80	AI in Natural Language Processing	Advancements and challenges in NLP applications
81	AI in Computer Vision	The role of AI in interpreting visual information
82	AI in Robotics	Integrating AI for intelligent robotic systems
83	AI in Autonomous Vehicles	Applications and considerations in self-driving cars
84	AI in Gaming	Enhancing gaming experiences with AI technologies
85	AI in Cybersecurity	Leveraging AI for threat detection and prevention
86	AI in Agriculture	Precision farming and crop management with AI
87	AI in Retail	Applications of AI in the retail industry
88	AI in Marketing	Personalization and analytics in marketing with AI
89	AI in Education	Transformative applications in the education sector
90	AI in Human Resources	Improving HR processes with AI-driven solutions
91	AI in Climate Science	Contributions of AI to climate research and modeling
92	AI in Space Exploration	Enhancing space missions with AI technologies
93	AI in Art and Creativity	Creative applications and generative art with AI
94	AI in Music	Composing, producing, and enhancing music with AI
95	AI in Sports	Performance analysis, injury prediction, and fan engagement
96	AI in Language Translation	Improving language translation with neural networks
97	AI in Virtual Assistants	Advancements in intelligent virtual assistant technologies
98	AI in Biotechnology	Applications of AI in biopharmaceutical research
99	AI in Telecommunications	Enhancing network management and communication
100	AI in Law	Applications of AI in the legal industry

Table 2: Fairness and Ethics in AI

No.	Topic	Content	LINK
1	Bias in AI	Understanding types of bias in AI and methods of identification
2	Fair AI	Principles and technologies for creating fair AI models
3	Explainable AI (XAI)	Understanding the importance and methods of model interpretability
4	Fairness Indicators	Understanding metrics for measuring model fairness
5	Counterfactual Fairness	Concept and application of counterfactual fairness
6	Adversarial Debiasing	Understanding bias removal methods through adversarial training
7	Data Preprocessing	Understanding bias removal methods through data preprocessing
8	A/B Testing	Understanding methods for evaluating and comparing model performance
9	Model Explainability	Understanding methods for explaining model predictions and decisions
10	SHAP Values	Meaning and usage of SHapley Additive exPlanations
11	LIME	Concept and usage of Local Interpretable Model-agnostic Explanations
12	Sensitivity Analysis	Understanding methods for analyzing output changes with input variations
13	Data Bias	Understanding causes and solutions for bias in data
14	Demographic Parity	Understanding fairness metrics related to demographic characteristics
15	Equalized Odds	Understanding fairness metric ensuring equal prediction errors between groups
16	Group Fairness	Understanding methods for maintaining fairness between groups
17	Individual Fairness	Understanding methods for maintaining fairness at the individual level
18	Intersectional Fairness	Understanding fairness considering various characteristics
19	Model Evaluation	Understanding quantitative methods for evaluating model performance
20	ROC-AUC	Understanding the meaning and usage of ROC-AUC
21	Precision-Recall Curve	Understanding the meaning and interpretation of precision-recall curves
22	Bias Mitigation	Understanding types of bias mitigation methods and their applications
23	Fairness-aware Machine Learning	Understanding approaches to machine learning considering fairness
24	Counterfactual Explanations	Understanding model interpretation through counterfactual explanations
25	Data Sampling	Understanding bias control methods through data sampling
26	Adversarial Training	Understanding model robustness through adversarial training
27	Responsible AI	Understanding ethical responsibilities and principles in AI usage
28	Interpretability vs. Explainability	Understanding the difference between interpretability and explainability
29	Privacy-preserving AI	Understanding methods for developing AI models with privacy preservation
30	Algorithmic Bias	Understanding types of bias in algorithms and methods for resolution
31	Causal Inference	Understanding methods for inferring causation and its importance
32	Feature Importance	Understanding methods for evaluating the importance of features
33	Bias Detection	Understanding methods and tools for detecting bias
34	Debiased Embeddings	Understanding methods for removing bias in embeddings
35	Recourse in ML	Understanding possibilities for improving model errors
36	Human-in-the-loop AI	Understanding methods for model improvement through human judgment
37	Ethical AI	Understanding principles for the ethical development of AI models
38	Data Governance	Understanding principles and importance of data governance
39	Data Ethics	Understanding ethical responsibilities in data collection, storage, and usage
40	Data Privacy Regulations	Understanding regulations for data privacy
41	Fairness-aware Algorithms	Understanding various algorithms considering fairness
42	Fairness-aware Optimization	Understanding methods for improving fairness in optimization
43	Fairness-aware Representation Learning	Understanding methods for improving fairness in representation learning
44	Fairness-aware Clustering	Understanding methods for maintaining fairness in clustering
45	Responsible Data Science	Understanding ethical responsibilities in data science
46	Responsible AI Governance	Understanding responsible operations for AI systems
47	Ethical AI Decision-making	Understanding AI model development with ethical decision-making
48	AI for Social Good	Understanding AI's societal responsibility and contribution
49	Data Bias Mitigation	Understanding methods for reducing bias in data
50	Fairness-aware Data Collection	Understanding methods for maintaining fairness in data collection