# Reinforcement Learning Enhancement Using Vector Semantic Representation and Symbolic Reasoning for Human-Centered Autonomous Emergency Braking

arXiv: https://arxiv.org/abs/2602.05079 · HTML: https://mlsanity.com/paper/2602.05079/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning.
- Builds on 16 of its 43 references that are in the map.
- First posted on arXiv 4 Feb 2026.

## Placed in

- LLM reasoning / [Chain-of-thought reasoning for driving](https://mlsanity.com/lr/chain-of-thought-reasoning-for-driving/)

## Builds on (most cited first, 16 of 16)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving](https://mlsanity.com/paper/2503.07656/) (2025)
- [Enhance Sample Efficiency and Robustness of End-to-end Urban Autonomous Driving via Semantic Masked World Model](https://mlsanity.com/paper/2210.04017/) (2022)
- [Explaining Autonomous Driving by Learning End-to-End Visual Attention](https://mlsanity.com/paper/2006.03347/) (2020)
- [Learning with Holographic Reduced Representations](https://mlsanity.com/paper/2109.02157/) (2021)
- [End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanations](https://mlsanity.com/paper/2403.12451/) (2024)
- [PyReason: Software for Open World Temporal Logic](https://mlsanity.com/paper/2302.13482/) (2023)
- [AUTO-DISCERN: Autonomous Driving Using Common Sense Reasoning](https://mlsanity.com/paper/2110.13606/) (2021)
- [Recasting Self-Attention with Holographic Reduced Representations](https://mlsanity.com/paper/2305.19534/) (2023)
- [LogiCity: Advancing Neuro-Symbolic AI with Abstract Urban Simulation](https://mlsanity.com/paper/2411.00773/) (2024)
- [Towards Safe Autonomous Driving Policies using a Neuro-Symbolic Deep Reinforcement Learning Approach](https://mlsanity.com/paper/2307.01316/) (2023)
- [Gluing Neural Networks Symbolically Through Hyperdimensional Computing](https://mlsanity.com/paper/2205.15534/) (2022)
- [Efficient Off-Policy Reinforcement Learning via Brain-Inspired Computing](https://mlsanity.com/paper/2205.06978/) (2022)
- [Offline Reinforcement Learning using Human-Aligned Reward Labeling for Autonomous Emergency Braking in Occluded Pedestrian Crossing](https://mlsanity.com/paper/2504.08704/) (2025)

## Abstract (arXiv)

The problem with existing camera-based Deep Reinforcement Learning approaches is twofold: they rarely integrate high-level scene context into the feature representation, and they rely on rigid, fixed reward functions. To address these challenges, this paper proposes a novel pipeline that produces a neuro-symbolic feature representation that encompasses semantic, spatial, and shape information, as well as spatially boosted features of dynamic entities in the scene, with an emphasis on safety-critical road users. It also proposes a Soft First-Order Logic (SFOL) reward function that balances human values via a symbolic reasoning module. Here, semantic and spatial predicates are extracted from segmentation maps and applied to linguistic rules to obtain reward weights. Quantitative experiments in the CARLA simulation environment show that the proposed neuro-symbolic representation and SFOL reward function improved policy robustness and safety-related performance metrics compared to baseline representations and reward formulations across varying traffic densities and occlusion levels. The findings demonstrate that integrating holistic representations and soft reasoning into Reinforcement Learning can support more context-aware and value-aligned decision-making for autonomous driving.