# Tiny, On-Device Decision Makers with the MiniConv Library

arXiv: https://arxiv.org/abs/2512.19726 · HTML: https://mlsanity.com/paper/2512.19726/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 16 of its 34 references that are in the map.
- First posted on arXiv 17 Dec 2025.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

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

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Deep Reinforcement Learning for Autonomous Driving: A Survey](https://mlsanity.com/paper/2002.00444/) (2020)
- [Learning Dexterous In-Hand Manipulation](https://mlsanity.com/paper/1808.00177/) (2018)
- [To prune, or not to prune: exploring the efficacy of pruning for model compression](https://mlsanity.com/paper/1710.01878/) (2017)
- [What is the State of Neural Network Pruning?](https://mlsanity.com/paper/2003.03033/) (2020)
- [AMC: AutoML for Model Compression and Acceleration on Mobile Devices](https://mlsanity.com/paper/1802.03494/) (2018)
- [Sim-to-Real Reinforcement Learning for Deformable Object Manipulation](https://mlsanity.com/paper/1806.07851/) (2018)
- [Distributed Deep Neural Networks over the Cloud, the Edge and End Devices](https://mlsanity.com/paper/1709.01921/) (2017)
- [Virtual-to-real Deep Reinforcement Learning: Continuous Control of Mobile Robots for Mapless Navigation](https://mlsanity.com/paper/1703.00420/) (2017)
- [Edge Intelligence: On-Demand Deep Learning Model Co-Inference with Device-Edge Synergy](https://mlsanity.com/paper/1806.07840/) (2018)
- [Generalization through Simulation: Integrating Simulated and Real Data into Deep Reinforcement Learning for Vision-Based Autonomous Flight](https://mlsanity.com/paper/1902.03701/) (2019)
- [Quantisation and Pruning for Neural Network Compression and Regularisation](https://mlsanity.com/paper/2001.04850/) (2020)

## Built on by (most cited first, 1 of 1)

- [BitRL: Reinforcement Learning with 1-bit Quantized Language Models for Resource-Constrained Edge Deployment](https://mlsanity.com/paper/2604.24273/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) has achieved strong results, but deploying visual policies on resource-constrained edge devices remains challenging due to computational cost and communication latency. Many deployments therefore offload policy inference to a remote server, incurring network round trips and requiring transmission of high-dimensional observations. We introduce a split-policy architecture in which a small on-device encoder, implemented as OpenGL fragment-shader passes for broad embedded GPU support, transforms each observation into a compact feature tensor that is transmitted to a remote policy head. In RL, this communication overhead manifests as closed-loop decision latency rather than only per-request inference latency. The proposed approach reduces transmitted data, lowers decision latency in bandwidth-limited settings, and reduces server-side compute per request, whilst achieving broadly comparable learning performance by final return (mean over the final 100 episodes) in single-run benchmarks, with modest trade-offs in mean return. We evaluate across an NVIDIA Jetson Nano, a Raspberry Pi 4B, and a Raspberry Pi Zero 2 W, reporting learning results, on-device execution behaviour under sustained load, and end-to-end decision latency and scalability measurements under bandwidth shaping. Code for training, deployment, and measurement is released as open source.