Suppose we define a 2-order stress tensor \(P_{ij}\), where \(i\) stands for scalar components constituting the vector and \(j\) represents the dimension. Giving a specific normalized direction \(n...

Why Lagrange multiplier \(\lambda\) can work Let’s consider general multi-dimensional constrained optimization problems. To find the optimized point, one can start from where the constraint is sa...

Math derivation for a CNN problem Let’s consider the CNN operation: [\begin{aligned} \sum_j f_1(x) =& \sum_j \alpha_j A_j(x,x_j) := \sum_j \alpha_j A_j \sum_j f_2(x) =& \sum_j \beta_j B_...

Deep Learning Framework for Physics-based Cloth Simulation Physics-embedded NN structure for machine learning application in Computer Graphics cloth animation. Direct PBS features are encoded into...

SPACE STYLE DRAGON SCENE ANIMATION A WebGL-based Chinese dragon animation with STL loader and GLSL shader imposed via Three.js. Project Link on GitHub Table of Contents Features Dependency...

Side Tips Insert below code into _layouts/default.html or _layouts/post.html to enable LaTex writing in Jekyll <script src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AM...

Deep-reinforcement-learning Controlled Fluid-Rigid Simulation A python-based ball-shooting game of fluid-rigid simulation and autonomous control via Deep Reinforcement Learning. Project Link on G...

1 Policy Evaluation 1.1 Monte-Carlo Estimation Recall that \(R_t(h)=R(s,a)|_t+\gamma\cdot R(s,a)|_{t+1}+\dots+{\gamma}^{H-t-1}\cdot R(s,a)|_{H-1}+{\gamma}^{H-t}R(s,a)|_H\) \(~~~~~~~~~~=\sum_{\tau...

GL Rendering A c++ OpenGL rendering practice, including various shaders, GLSL programming, illumination models; transparency, shadows; modeling, heterogeneous textures; camera and motion. Project...

Self-built Ubuntu Server You might as well refer the link BV1554y1n7zv, but below are some notifications: 1.Enter the installation through Safe GUI 1.When parting the disk, use four partitions: ...