IA et Machine Learning
Machine learning, génération augmentée par récupération (RAG), applications d'IA et web crawling.
Advancing Vapor Pressure Prediction: A Machine Learning Approach with Directed Message Passing Neural Networks
12 December 2024 Advancing Vapor Pressure Prediction: A Machine Learning Approach with Directed Message Passing Neural Networks Yen-Hsiang Lin1, Hsin-Hao Liang1, Shiang-Tai Lin1, Yi-Pei Li1 1. National Taiwan University
Lire l'article →Hybrid Neural Networks for Improved Chemical
1 Hybrid Neural Networks for Improved Chemical Process Modeling: Bridging Data-Driven Insights with Physical Consistency Jana Mousa *, Stéphane Negny, Rachid Ouaret. Laboratoire de Génie Chimique, Université de Toulouse,
Lire l'article →Leveraging spatial charge descriptor in deep learning models: Toward highly accurate prediction of vapor-liquid equilibrium
Leveraging spatial charge descriptor in deep learning models: Toward highly accurate prediction of vapor-liquid equilibrium Hsiu-Min Hung , Ying-Chieh Hung * Department of Chemical Engineering and Biotechnology, National
Lire l'article →00 overview
OVERVIEW Machine Learning & Deep Learning A depth-first curriculum, from math foundations to production Phase goal Turn “I can call .fit()” into “I understand, can implement, and can ship. ” This guide is the map for the
Lire l'article →1 1 Regression
ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2 ˆy = β⊤x R2REGRESSION FROM BEGINNER TO EXPERT A complete, visual, application-driven course — from fitting a line through po
Lire l'article →1 2 Bayes
P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H)BAYESIAN INFE
Lire l'article →1 3 Tree Based Methods
xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? TREE-BASED METHODS FROM BEGINNER TO EXPERT A complete, visual, application-driven course — from a single decision tree through ran- dom forests and
Lire l'article →1 4 Trees
xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? xj < t ? TREE-BASED METHODS FROM BEGINNER TO EXPERT A complete, visual, application-driven course — from a single decision tree through ran- dom forests and
Lire l'article →1 6 Support Vector Machines
w⊤x + b w⊤x + b w⊤x + b w⊤x + b w⊤x + b w⊤x + b w⊤x + b w⊤x + b SUPPORT VECTOR MACHINES FROM BEGINNER TO EXPERT A complete, visual, application-driven course — from the maximal-margin idea through soft margins, the kerne
Lire l'article →Calculus
Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f Z dy dx limh→0 ∇f CALCULUS FROM BEGINNER TO EXPERT A compl
Lire l'article →Linear Algebra
a00 a01 a02 a03 a04 a05 a06 a07 a08 a09 a010 a011 a012 a013 a014 a015 a016 a017 a018 a019 a020 a10 a11 a12 a13 a14 a15 a16 a17 a18 a19 a110 a111 a112 a113 a114 a115 a116 a117 a118 a119 a120 a20 a21 a22 a23 a24 a25 a26 a2
Lire l'article →ML DL Curriculum complete
OVERVIEW Machine Learning & Deep Learning A depth-first curriculum, from math foundations to production Phase goal Turn “I can call .fit()” into “I understand, can implement, and can ship. ” This guide is the map for the
Lire l'article →Probability & Statistics
P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] P (A | B) µ, σ2 E[X] PROBABILITY & STATISTICS FROM B
Lire l'article →bayes
P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H) P (H| D) ∝ P (D|H ) P (H)BAYESIAN INFE
Lire l'article →linear algebra
a00 a01 a02 a03 a04 a05 a06 a07 a08 a09 a010 a011 a012 a013 a014 a015 a016 a017 a018 a019 a020 a10 a11 a12 a13 a14 a15 a16 a17 a18 a19 a110 a111 a112 a113 a114 a115 a116 a117 a118 a119 a120 a20 a21 a22 a23 a24 a25 a26 a2
Lire l'article →phase0 math tooling
PHASE 0 Math & Tooling Foundations Linear algebra, calculus, probability, statistics, and the scientific Python stack Phase goal Build the mathematical intuition that makes every later algorithm click instead of feeling
Lire l'article →phase1 core ml
PHASE 1 Core Machine Learning Classical models, evaluation, feature engineering, and tuning — implemented, not just imported Phase goal Understand classical ML deeply enough to implement the key algorithms from scratch a
Lire l'article →phase2 dl foundations
PHASE 2 Deep Learning Foundations Neurons, backpropagation, optimization, regularization, and PyTorch — from first prin- ciples Phase goal Understand neural networks from the ground up — forward pass, backpropagation, op
Lire l'article →phase3 computer vision
PHASE 3 Computer Vision Convolutions, CNN architectures, transfer learning, detection, segmentation, and ViTs Phase goal Understand how neural networks see: the convolution operation, the architectural lineage from LeNet
Lire l'article →phase4 nlp transformers
PHASE 4 Sequence Models, NLP & Transform-ers From RNNs to attention to LLMs, tokenization, fine-tuning, and RAG Phase goal Master the architecture behind modern AI. Build a Transformer from first princi- ples, understand
Lire l'article →phase5 generative models
PHASE 5 Generative Models Autoencoders, VAEs, GANs, and diffusion — how machines create Phase goal Understand how models learn to generate data: the latent-variable view (V AEs), the adversarial game (GANs), and the deno
Lire l'article →phase6 reinforcement learning
PHASE 6 Reinforcement Learning MDPs, value functions, Q-learning, policy gradients, PPO, and RLHF Phase goal Understand learning by interaction and reward: Markov decision processes, the value/policy duality, deep Q-netw
Lire l'article →phase7 mlops deployment
PHASE 7 MLOps, Deployment & Production Systems Experiment tracking, serving, containers, monitoring, optimization, and CI/CD for ML Phase goal Turn models into reliable products. Track experiments, package and serve mode
Lire l'article →phase8 capstones
PHASE 8 Capstone Projects Compounding value by building end-to-end systems in your own domains Phase goal Convert knowledge into demonstrable capability. Ship 2–3 end-to-end projects that intersect your existing domains
Lire l'article →