Clik here to view.
Review sách: The Master Algorithm
Dạo gần đây rảnh nên đọc được nhiều hơn, tạm thời sẽ review sách trước khi quay lại với series về Học củng cố (a.k.a. Reinforcement Learning). Mình biết tới Pedro Domingos cách đây vài năm, khi đang...
View ArticleClik here to view.
Deep Generative models : Recent advances – part 1
Một cách khá “mỉa mai”, thành công của Deep Learning gần đây chủ yếu tập trung trong các bài toán supervised, trong đó các mô hình Deep Learning chủ yếu học phân phối xác suất có điều kiện của label...
View ArticleClik here to view.
Bayesian Optimization part 1: Gaussians and Gaussian Process
SigOpt, công ty làm về Bayesian Optimization mà có lần mình làm chung vài thứ linh tinh, có hẳn một booth trong KDD năm nay. Nói như Nando de Freitas: “Bayesian Optimization is a thing“, thành ra mình...
View ArticleClik here to view.
[RL3a] Reinforcement Learning context
[RL1] Markov Decision process – Introduction [RL2a] Markov Decision Process – Discounted Reward [RL2b] Markov Decision Process – Bellman equation [RL2c] Markov Decision Process – Solving Bellman...
View ArticleClik here to view.
[RL3b] Temporal Difference Learning – intuition
[RL1] Markov Decision process – Introduction [RL2a] Markov Decision Process – Discounted Reward [RL2b] Markov Decision Process – Bellman equation [RL2c] Markov Decision Process – Solving Bellman...
View ArticleClik here to view.
Variational Autoencoders 2: Maths
Variational Autoencoders 1: Overview Variational Autoencoders 2: Maths Variational Autoencoders 3: Training, Inference and comparison with other models Last time we saw the probability distribution of...
View ArticleClik here to view.
Variational Autoencoders 3: Training, Inference and comparison with other models
Variational Autoencoders 1: Overview Variational Autoencoders 2: Maths Variational Autoencoders 3: Training, Inference and comparison with other models Recalling that the backbone of VAEs is the...
View ArticleClik here to view.
Kalman filters (and how they relate to HMMs)
Kalman filters are insanely popular in many engineering fields, especially those involve sensors and motion tracking. Consider how to design a radar system to track military aircrafts (or warships,...
View ArticleClik here to view.
[RL4a] Policy Optimization
I thought I would write about some theory behind Reinforcement Learning, with eligibility traces, contraction mapping, POMDP and so on, but then I realized if I go down that rabbit hole, I would...
View ArticleClik here to view.
Simpson’s paradox
I learned about the Simpson’s paradox fairly recently, and I found it quite disturbing, not because of the mere “paradox” itself, but mainly because I felt it was something I should have known already....
View Article