Design and Implementation of a High‑Scoring Tetris AI for the Tencent Geek Challenge

Zheng Lin‑kai’s record‑breaking Tetris AI for the Tencent Geek Challenge combines a two‑layer breadth‑first search—stage‑level pruning in Python and fast round‑level BFS in C++—with a heuristic that rewards high score, low occupied cells, and smooth board transitions, enabling a 1,413,876‑point performance.

C++Monte Carlo Tree SearchPython

0 likes · 10 min read

Design and Implementation of a High‑Scoring Tetris AI for the Tencent Geek Challenge

Tencent Cloud Developer

Aug 20, 2021 · Artificial Intelligence

How I Scored 1.38 Million in Tencent’s Tetris Challenge with AI and Dynamic Programming

This article details a graduate student's AI‑driven solution to the Tencent Geek Competition's "E‑Tetris" challenge, describing the EI‑Tetris heuristic, feature‑based evaluation, state‑pruned dynamic programming, C++ implementation, performance results, and potential improvements.

Evaluation FunctionState PruningTetris AI

0 likes · 12 min read

How I Scored 1.38 Million in Tencent’s Tetris Challenge with AI and Dynamic Programming

Design and Implementation of a High‑Scoring Tetris AI for the Tencent Geek Challenge

How I Scored 1.38 Million in Tencent’s Tetris Challenge with AI and Dynamic Programming

How I Scored 1.38 Million in Tencent’s Tetris Challenge with AI and Dynamic Programming