---
title: "Using Transformers for Teaching and Learning via Symbolic Distillation of Neural A-life Agents"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.07254v1"
permalink: "https://feed.astralor.com/items/itm_8019af235e0073530464be46"
published: "2026-11-05T00:06:00.000Z"
score: 58
channel: "ai"
tags: ["AI Agent","Using Transformers","Teaching","Learning","Symbolic Distillation","Neural A-life Agents","Neural","A-life"]
---

# Using Transformers for Teaching and Learning via Symbolic Distillation of Neural A-life Agents

## 速览

Neural networks controlling behavior of A-life agents can evolve over multiple generations. While effective in making the agents survive, such networks are generally opaque to a human observer. Furthermore, the neural w…

## 关键点

- Neural networks controlling behavior of A-life agents can evolve over multiple generations.
- While effective in making the agents survive, such networks are generally opaque to a human observer.
- Furthermore, the neural weights cannot be transferred between agents with different neural architectures.

## 全文

Neural networks controlling behavior of A-life agents can evolve over multiple generations. While effective in making the agents survive, such networks are generally opaque to a human observer. Furthermore, the neural weights cannot be transferred between agents with different neural architectures. Distilling such neural networks into compact symbolic policies can help humans understand agent behavior in A-life. Additionally, it allows the agents to share their knowledge in a compact fashion, improving population-level learning. Distilling a neural policy into a symbolic policy is typically performed by searching the space of symbolic policies to find one that approximates the neural policy. However, even with a domain-specific language the space of policies can be large, necessitating incomplete stochastic search such as genetic algorithms. Yet the distillation remains computationally expensive. We propose to instead use a transformer pre-trained on readily available random percept-action data. Doing so speeds up distillation substantially. We derive the transformer architecture via a neural architecture search. Empirical results compare different distillation methods and the influence of the symbolic space on A-life population dynamics through self-reflection and teaching and learning.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_8019af235e0073530464be46