# \[Paper\] Neural circuit policies enabling auditable autonomy

**URL:** <https://discourse.numenta.org/t/paper-neural-circuit-policies-enabling-auditable-autonomy/7945>\
**Category:** Related Papers\
**Created:** [October 14, 2020, 4:14pm UTC](https://discourse.numenta.org/t/paper-neural-circuit-policies-enabling-auditable-autonomy/7945 "2020-10-14T16:14:00Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![LuizCoutinho](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/luizcoutinho/32/3703_2.png) [@LuizCoutinho](https://discourse.numenta.org/u/LuizCoutinho)\
**Post date:** [October 14, 2020, 4:14pm UTC](https://discourse.numenta.org/t/paper-neural-circuit-policies-enabling-auditable-autonomy/7945/1 "2020-10-14T16:14:01Z")

</div>

Abstract  
A central goal of artificial intelligence in high-stakes decision-making applications is to design a single algorithm that simultaneously expresses generalizability by learning coherent representations of their world and interpretable explanations of its dynamics. Here, we combine brain-inspired neural computation principles and scalable deep learning architectures to design compact neural controllers for task-specific compartments of a full-stack autonomous vehicle control system. We discover that a single algorithm with 19 control neurons, connecting 32 encapsulated input features to outputs by 253 synapses, learns to map high-dimensional inputs into steering commands. This system shows superior generalizability, interpretability and robustness compared with orders-of-magnitude larger black-box learning systems. The obtained neural agents enable high-fidelity autonomy for task-specific parts of a complex autonomous system.

[https://www.nature.com/articles/s42256-020-00237-3](https://www.nature.com/articles/s42256-020-00237-3)

[PDF](https://www.nature.com/articles/s42256-020-00237-3.epdf?sharing_token=xHsXBg2SoR9l8XdbXeGSqtRgN0jAjWel9jnR3ZoTv0PbS_e49wmlSXvnXIRQ7wyir5MOFK7XBfQ8sxCtVjc7zD1lWeQB5kHoRr4BAmDEU0_1-UN5qHD5nXYVQyq5BrRV_tFa3_FZjs4LBHt-yebsG4eQcOnNsG4BenK3CmBRFLk%3D)

---

<div class="post-metadata">

**Author:** ![cezar\_t](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/cezar_t/32/4451_2.png) [@cezar\_t](https://discourse.numenta.org/u/cezar_t)\
**Post date:** [October 16, 2020, 10:10pm UTC](https://discourse.numenta.org/t/paper-neural-circuit-policies-enabling-auditable-autonomy/7945/2 "2020-10-16T22:10:43Z")

</div>

[![](https://img.youtube.com/vi/wAa358pNDkQ/maxresdefault.jpg "A new brain-inspired intelligent system can drive a car using only 19 control neurons!") ](https://www.youtube.com/watch?v=wAa358pNDkQ)

Till someone solves human brain, (a model of) the nematode’s seems sufficient to keep the lane.

---

<div class="post-metadata">

**Author:** ![SeanOConnor](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/seanoconnor/32/3441_2.png) [@SeanOConnor](https://discourse.numenta.org/u/SeanOConnor)\
**Post date:** [October 15, 2021, 12:18am UTC](https://discourse.numenta.org/t/paper-neural-circuit-policies-enabling-auditable-autonomy/7945/3 "2021-10-15T00:18:30Z")

</div>

There is a cause and effect article on Neural Circuit Policies here:  
[https://techxplore.com/news/2021-10-ai-agent-cause-and-effect-basis-task.html](https://techxplore.com/news/2021-10-ai-agent-cause-and-effect-basis-task.html)

[https://arxiv.org/abs/1803.08554](https://arxiv.org/abs/1803.08554)

I’ll read up on it a bit. Self modifying code was a thing in the 1980s before pipelined CPUs, you can make self modifying neural networks.