Luna's Blog

「恰似天地一蜉蝣」

CPSC 340 Feature Selection

Feature Selection General feature selection problem: Find the features (columns) of ‘X’ that are important for predicting ‘y’. Association Approach: For each feature ‘j’, compute correlation be...

Posted by Luna on March 7, 2022

Trees

Bottom Up Recursive VS Top Down Recursive Ancestor Problems 235. Lowest Common Ancestor of a Binary Search Tree 236. Lowest Common Ancestor of a Binary Tree 1026. Maximu...

Posted by Luna's Blog on February 26, 2022

Math Problems

50. Pow(x, n) Implement pow(x, n), which calculates x raised to the power n (i.e., $x^n$). The simplest way is to multiply n times, time complexity is $O(n)$ Another way is to think $x^{10}$ as $...

Posted by Luna's Blog on February 26, 2022

SweepLine Related Problems

Interval & Sweep Line Related Problems For two intervals, A and B: if A.end > B.start and A.start < B.end then they have duplicate intervals. After sorting by the start time, w...

Posted by Luna's Blog on February 17, 2022

CPSC 340 Robust Regression

Robust Regression Robust regression focus less on large errors. Instead of minimizing L2-norm, minimizes L1-norm of residuals. $f(w) = ||Xw-y||$ – it’s harder to minimize the absolute error, we d...

Posted by Luna's Blog on February 16, 2022

CPSC 340 Feature Selection

We need to find the features that are important for predicting y. Association approach - for each feature j, compute correlation between feature $x^j$ and $y$ (however it ignores variable inter...

Posted by Luna's Blog on February 16, 2022

Linked List

Cycles We can use Floyd algorithm to solve problems like this. 141. Linked List Cycle Given head, the head of a linked list, determine if the linked list has a cycle in it. Return true if there...

Posted by Luna's Blog on February 15, 2022

CPSC 340 Linear Regression

We want to discover relationship between numerical variables. Linear Regression in 1-D dimension Assume we have only 1 feature (d = 1), linear regression make prediction $\hat{y_i}$ using a linear...

Posted by Luna's Blog on February 14, 2022

CPSC 340 Gradient Descent

Gradient Descent starts with a guess $w_0$ use the gradient $\nabla f(w^0)$ to generate a better $w^1$ … the limit of $w^t$ goes to $\infty$ has $\nabla f(w^t) = 0$ It converges to a glo...

Posted by Luna's Blog on February 14, 2022

CPSC340 Outlier Detection

Methods Model-based Graphical Cluster-Based Distance-Based Supervised Model-Based Outlier Detection We can fit a probabilistic model, outliers are those with low probability. Assume d...

Posted by Luna's Blog on February 13, 2022

FEATURED TAGS

Japanese algorithm cpsc340 leetcode machine learning ubc frontend binary search

ABOUT ME

陆 · 🐱之使徒 · 🏂🏻摔跤者 · 💻 · 蕾
Lu, Lei (Luna)
🇨🇳 ➡️ 🇨🇦
🐱 🏂🏻 💻