# Policy Distillation with Selective Input Gradient Regularization for Efficient Interpretability

arXiv: https://arxiv.org/abs/2205.08685 · HTML: https://mlsanity.com/paper/2205.08685/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 18 May 2022.

## Builds on (most cited first, 24 of 24)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [SmoothGrad: removing noise by adding noise](https://mlsanity.com/paper/1706.03825/) (2017)
- [Striving for Simplicity: The All Convolutional Net](https://mlsanity.com/paper/1412.6806/) (2014)
- [Dueling Network Architectures for Deep Reinforcement Learning](https://mlsanity.com/paper/1511.06581/) (2015)
- [Interpretable Explanations of Black Boxes by Meaningful Perturbation](https://mlsanity.com/paper/1704.03296/) (2017)
- [Top-down Neural Attention by Excitation Backprop](https://mlsanity.com/paper/1608.00507/) (2016)
- [Adversarial Attacks on Neural Network Policies](https://mlsanity.com/paper/1702.02284/) (2017)
- [Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations](https://mlsanity.com/paper/2003.08938/) (2020)
- [Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients](https://mlsanity.com/paper/1711.09404/) (2017)
- [Visualizing and Understanding Atari Agents](https://mlsanity.com/paper/1711.00138/) (2017)
- [Distilling Policy Distillation](https://mlsanity.com/paper/1902.02186/) (2019)
- [Explain Your Move: Understanding Agent Actions Using Specific and Relevant Feature Attribution](https://mlsanity.com/paper/1912.12191/) (2019)
- [Transparency and Explanation in Deep Reinforcement Learning Neural Networks](https://mlsanity.com/paper/1809.06061/) (2018)
- [Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samples](https://mlsanity.com/paper/1810.11580/) (2018)
- [Robust Deep Reinforcement Learning for Security and Safety in Autonomous Vehicle Systems](https://mlsanity.com/paper/1805.00983/) (2018)
- [Are Gradient-based Saliency Maps Useful in Deep Reinforcement Learning?](https://mlsanity.com/paper/2012.01281/) (2020)

## Built on by (most cited first, 2 of 2)

- [Jacobian Norm with Selective Input Gradient Regularization for Improved and Interpretable Adversarial Defense](https://mlsanity.com/paper/2207.13036/) (2022)
- [MSVIPER: Improved Policy Distillation for Reinforcement-Learning-Based Robot Navigation](https://mlsanity.com/paper/2209.09079/) (2022)

## Abstract (arXiv)

Although deep Reinforcement Learning (RL) has proven successful in a wide range of tasks, one challenge it faces is interpretability when applied to real-world problems. Saliency maps are frequently used to provide interpretability for deep neural networks. However, in the RL domain, existing saliency map approaches are either computationally expensive and thus cannot satisfy the real-time requirement of real-world scenarios or cannot produce interpretable saliency maps for RL policies. In this work, we propose an approach of Distillation with selective Input Gradient Regularization (DIGR) which uses policy distillation and input gradient regularization to produce new policies that achieve both high interpretability and computation efficiency in generating saliency maps. Our approach is also found to improve the robustness of RL policies to multiple adversarial attacks. We conduct experiments on three tasks, MiniGrid (Fetch Object), Atari (Breakout) and CARLA Autonomous Driving, to demonstrate the importance and effectiveness of our approach.