# Connecting Concept Convexity and Human-Machine Alignment in Deep Neural Networks

arXiv: https://arxiv.org/abs/2409.06362 · HTML: https://mlsanity.com/paper/2409.06362/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 26 references that are in the map.
- First posted on arXiv 10 Sep 2024.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [Layer-wise Analysis of a Self-supervised Speech Representation Model](https://mlsanity.com/paper/2107.04734/) (2021)
- [AI Alignment: A Comprehensive Survey](https://mlsanity.com/paper/2310.19852/) (2023)
- [Comparative layer-wise analysis of self-supervised speech models](https://mlsanity.com/paper/2211.03929/) (2022)
- [Getting aligned on representational alignment](https://mlsanity.com/paper/2310.13018/) (2023)
- [Human alignment of neural network representations](https://mlsanity.com/paper/2211.01201/) (2022)
- [Harmonizing the object recognition strategies of deep neural networks with humans](https://mlsanity.com/paper/2211.04533/) (2022)
- [Improving neural network representations using human similarity judgments](https://mlsanity.com/paper/2306.04507/) (2023)
- [Alignment with human representations supports robust few-shot learning](https://mlsanity.com/paper/2301.11990/) (2023)
- [Transforming Neural Network Visual Representations to Predict Human Judgments of Similarity](https://mlsanity.com/paper/2010.06512/) (2020)
- [VICE: Variational Interpretable Concept Embeddings](https://mlsanity.com/paper/2205.00756/) (2022)
- [Passive Attention in Artificial Neural Networks Predicts Human Visual Selectivity](https://mlsanity.com/paper/2107.07013/) (2021)
- [On convex decision regions in deep network representations](https://mlsanity.com/paper/2305.17154/) (2023)
- [How Aligned are Different Alignment Metrics?](https://mlsanity.com/paper/2407.07530/) (2024)
- [An Analysis of Human Alignment of Latent Diffusion Models](https://mlsanity.com/paper/2403.08469/) (2024)
- [Convexity in complex networks](https://mlsanity.com/paper/1608.03402/) (2016)
- [Convexity-based Pruning of Speech Representation Models](https://mlsanity.com/paper/2408.11858/) (2024)

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

- [Large Vision Models Can Solve Mental Rotation Problems](https://mlsanity.com/paper/2509.15271/) (2025)

## Abstract (arXiv)

Understanding how neural networks align with human cognitive processes is a crucial step toward developing more interpretable and reliable AI systems. Motivated by theories of human cognition, this study examines the relationship between convexity in neural network representations and human-machine alignment based on behavioral data. We identify a correlation between these two dimensions in pretrained and fine-tuned vision transformer models. Our findings suggest that the convex regions formed in latent spaces of neural networks to some extent align with human-defined categories and reflect the similarity relations humans use in cognitive tasks. While optimizing for alignment generally enhances convexity, increasing convexity through fine-tuning yields inconsistent effects on alignment, which suggests a complex relationship between the two. This study presents a first step toward understanding the relationship between the convexity of latent representations and human-machine alignment.