# Efficient Representation of the Activation Space in Deep Neural Networks

arXiv: https://arxiv.org/abs/2312.08143 · HTML: https://mlsanity.com/paper/2312.08143/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 37 references that are in the map.
- First posted on arXiv 13 Dec 2023.

## Builds on (most cited first, 19 of 19)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey](https://mlsanity.com/paper/1801.00553/) (2018)
- [Evaluating the visualization of what a Deep Neural Network has learned](https://mlsanity.com/paper/1509.06321/) (2015)
- [Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation](https://mlsanity.com/paper/1806.02473/) (2018)
- [Methods for Interpreting and Understanding Deep Neural Networks](https://mlsanity.com/paper/1706.07979/) (2017)
- [DeepXplore: Automated Whitebox Testing of Deep Learning Systems](https://mlsanity.com/paper/1705.06640/) (2017)
- [Detecting Adversarial Samples from Artifacts](https://mlsanity.com/paper/1703.00410/) (2017)
- [Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models](https://mlsanity.com/paper/1708.08296/) (2017)
- [DeepGauge: Multi-Granularity Testing Criteria for Deep Learning Systems](https://mlsanity.com/paper/1803.07519/) (2018)
- [FastDiff: A Fast Conditional Diffusion Model for High-Quality Speech Synthesis](https://mlsanity.com/paper/2204.09934/) (2022)
- [Interpretable Deep Learning: Interpretation, Interpretability, Trustworthiness, and Beyond](https://mlsanity.com/paper/2103.10689/) (2021)
- [Blacklight: Scalable Defense for Neural Networks against Query-Based Black-Box Attacks](https://mlsanity.com/paper/2006.14042/) (2020)
- [Towards creativity characterization of generative models via group-based subset scanning](https://mlsanity.com/paper/2104.00479/) (2021)
- [Out-of-Distribution Detection in Dermatology using Input Perturbation and Subset Scanning](https://mlsanity.com/paper/2105.11160/) (2021)
- [Identifying Audio Adversarial Examples via Anomalous Pattern Detection](https://mlsanity.com/paper/2002.05463/) (2020)
- [Pattern Detection in the Activation Space for Identifying Synthesized Content](https://mlsanity.com/paper/2105.12479/) (2021)
- [Subset Scanning Over Neural Network Activations](https://mlsanity.com/paper/1810.08676/) (2018)

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

- [One SPACE to Rule Them All: Jointly Mitigating Factuality and Faithfulness Hallucinations in LLMs](https://mlsanity.com/paper/2506.11088/) (2025)

## Abstract (arXiv)

The representations of the activation space of deep neural networks (DNNs) are widely utilized for tasks like natural language processing, anomaly detection and speech recognition. Due to the diverse nature of these tasks and the large size of DNNs, an efficient and task-independent representation of activations becomes crucial. Empirical p-values have been used to quantify the relative strength of an observed node activation compared to activations created by already-known inputs. Nonetheless, keeping raw data for these calculations increases memory resource consumption and raises privacy concerns. To this end, we propose a model-agnostic framework for creating representations of activations in DNNs using node-specific histograms to compute p-values of observed activations without retaining already-known inputs. Our proposed approach demonstrates promising potential when validated with multiple network architectures across various downstream tasks and compared with the kernel density estimates and brute-force empirical baselines. In addition, the framework reduces memory usage by 30% with up to 4 times faster p-value computing time while maintaining state of-the-art detection power in downstream tasks such as the detection of adversarial attacks and synthesized content. Moreover, as we do not persist raw data at inference time, we could potentially reduce susceptibility to attacks and privacy issues.