Posts – Reinforcement Learning/AI and Dynamic Programming/Control

Posts

Business projects are not available online because of Copyright and Business Secret.

Decentralized Optimal Control of Distributed Interdependent Automata With Priority Structure

Reinforcement Learning is Direct Adaptive Optimal Control

Neural-network-based decentralized control of continuous-time nonlinear interconnected systems with unknown dynamics

Decentralized Stabilization for a Class of Continuous-Time Nonlinear Interconnected Systems Using Online Learning Optimal Control Approach

Randomized Linear Programming Solves the Discounted Markov Decision Problem In Nearly-Linear (Sometimes Sublinear) Run Time

The Asymptotic Convergence-Rate of Q-learning

Solving H-horizon, Stationary Markov Decision Problems In Time Proportional To Log(H) Solving h-horizon, stationary markov decision problems in time proportional to log (h)

Finite-Sample Convergence Rates for Q-Learning and Indirect Algorithms

Hierarchical Apprenticeship Learning, with Application to Quadruped Locomotion

Policy Gradient Methods

Meta Learning Shared Hierarchies

Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation

Actor-Critic Algorithms for Hierarchical Markov Decision Processes

Hierarchical Policy Gradient Algorithms

Hierarchical Actor-Critic

Actor-Critic Algorithms

Policy Gradient Methods for Reinforcement Learning with Function Approximation

Reinforcement Learning with Soft State Aggregation

Markov Decision Processes and Policy Gradient

Policy Gradient and Q-learning

IntSim: A CAD tool for Optimization of Multilevel Interconnect Networks

Fourier Transform and Matrix-vector Multiplication

Euler Angles Bunge vs Elastic

Why do we use backpropagation?

Derivative of Sigmoid Function

Derivative of Hyperbolic Tangent Function

Compute Backpropagation Derivatives

Radix Sort
Bucket Sort
Counting Sort
Shell Sort
Heap Sort

Merge Sort
Quick Sort
Insertion Sort
The running time of quicksort depends on whether the partitioning is balanced or unbalanced,
which in turn depends on which elements are used for partitioning.
If the partitioning is balanced, the algorithm runs asymptotically as fast as merge sort.
If the partitioning is unbalanced, however, it can run asymptotically as slowly as insertion sort.
— <Introduction to Algorithms> 7.2 Performance of quicksort
Selection Sort
Bubble Sort

Remote access to the Linux Servers

Cadence Virtuoso

Elastic Tensor Analysis

Math Term Translate

Define Data Type in Matlab

OpenFOAM PDE Terms

OpenFOAM_Base_Units

OpenFOAM_transportProperties

Gradient MATLAB

Abnormal Login UESTC Alumni Email

The blogs below are about circuit design. Some figures from the posts below may be not available in Columbia Blog, as each blog has a limit of 50MB for uploaded files, e.g. images, with a per-file limit of 8MB in Columbia Blogs. Please visit https://peizhenlin.wordpress.com/ for posts with figures if you want to review them.

https://peizhenlin.wordpress.com/: home/ >ls

Configure STA environment

Timing verification

Crosstalk delay

Crosstalk delay on timing verification

Sequential cells timing models

Check clock gating

prepare files and check netlist

update clock latency

meet skew via constraint

set_disable_timing

Clock Tree Synthesis

Trace clock tree

Clock Tree Latency Skew Uncertainty

Clock divider and CTS

CTS Spec GlobalUnsyncPin RootPin based on Constraint and Netlist

Path delay in cross clock domain

Same source create_generated_clock -add

Linux/Unix file permissions

FinFET and MOSFET

Modeling_Power_Terminology