# Unsupervised Hebbian Learning on Point Sets in StarCraft II

arXiv: https://arxiv.org/abs/2207.12323 · HTML: https://mlsanity.com/paper/2207.12323/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 16 references that are in the map.
- First posted on arXiv 13 Jul 2022.

## Builds on (most cited first, 8 of 8)

- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [StarCraft II: A New Challenge for Reinforcement Learning](https://mlsanity.com/paper/1708.04782/) (2017)
- [Unsupervised Learning by Competing Hidden Units](https://mlsanity.com/paper/1806.10181/) (2018)
- [Hebbian Semi-Supervised Learning in a Sample Efficiency Setting](https://mlsanity.com/paper/2103.09002/) (2021)
- [DefogGAN: Predicting Hidden Information in the StarCraft Fog of War with Generative Adversarial Nets](https://mlsanity.com/paper/2003.01927/) (2020)
- [Deep Convolution for Irregularly Sampled Temporal Point Clouds](https://mlsanity.com/paper/2105.00137/) (2021)

## Built on by (most cited first, 3 of 3)

- [Unsupervised 3D Object Learning through Neuron Activity aware Plasticity](https://mlsanity.com/paper/2302.11622/) (2023)
- [Forecasting Evolution of Clusters in Game Agents with Hebbian Learning](https://mlsanity.com/paper/2209.06904/) (2022)
- [Forecasting Local Behavior of Self-organizing Many-agent System without Reconstruction](https://mlsanity.com/paper/2210.17289/) (2022)

## Abstract (arXiv)

Learning the evolution of real-time strategy (RTS) game is a challenging problem in artificial intelligent (AI) system. In this paper, we present a novel Hebbian learning method to extract the global feature of point sets in StarCraft II game units, and its application to predict the movement of the points. Our model includes encoder, LSTM, and decoder, and we train the encoder with the unsupervised learning method. We introduce the concept of neuron activity aware learning combined with k-Winner-Takes-All. The optimal value of neuron activity is mathematically derived, and experiments support the effectiveness of the concept over the downstream task. Our Hebbian learning rule benefits the prediction with lower loss compared to self-supervised learning. Also, our model significantly saves the computational cost such as activations and FLOPs compared to a frame-based approach.