Inherent Interpretability

Inherent interpretability in machine learning focuses on designing models and methods that are inherently transparent and understandable, aiming to reduce the "black box" nature of many AI systems. Current research emphasizes developing intrinsically interpretable model architectures, such as those based on decision trees, rule-based systems, and specific neural network designs (e.g., Kolmogorov-Arnold Networks), alongside techniques like feature attribution and visualization methods to enhance understanding of model behavior. This pursuit is crucial for building trust in AI, particularly in high-stakes applications like healthcare and finance, where understanding model decisions is paramount for responsible deployment and effective human-AI collaboration.

Papers

December 4, 2023

FeaInfNet: Diagnosis in Medical Image with Feature-Driven Inference and Visual Explanations
Yitao Peng, Lianghua He, Die Hu, Yihang Liu, Longzhen Yang, Shaohua Shang
Inherent Interpretability Medical Image Medical Diagnosis Visual Explanation Interpretable Deep Learning Machine Learning Inference Driven Inference

December 2, 2023

AS-XAI: Self-supervised Automatic Semantic Interpretation for CNN
Changqi Sun, Hao Xu, Yuntian Chen, Dongxiao Zhang
Inherent Interpretability xAI Community CNN Network Interpretable AI Semantic Model Global Interpretability Global Semantics

November 30, 2023

November 29, 2023

Interpreting Differentiable Latent States for Healthcare Time-series Data
Yu Chen, Nivedita Bijlani, Samaneh Kouchaki, Payam Barnaghi
Machine Learning Inherent Interpretability Latent State Healthcare Datasets Daily Activity

November 28, 2023

November 25, 2023

Everybody Needs a Little HELP: Explaining Graphs via Hierarchical Concepts
Jonas Jürß, Lucie Charlotte Magister, Pietro Barbiero, Pietro Liò, Nikola Simidjievski
Graph Drawing Inherent Interpretability Graph Pooling Social Network Analysis Hierarchical Concept Little Help

November 20, 2023

November 19, 2023

Rethinking Large Language Models in Mental Health Applications
Shaoxiong Ji, Tianlin Zhang, Kailai Yang, Sophia Ananiadou, Erik Cambria
Inherent Interpretability High Explainability Mental Health Self Explanation

November 13, 2023

Finding and Editing Multi-Modal Neurons in Pre-Trained Transformers
Haowen Pan, Yixin Cao, Xiaozhi Wang, Xun Yang, Meng Wang
Inherent Interpretability Multi Modal Large Language Model Pre Trained Transformer Cross Modal Representation Textual Representation

November 12, 2023

Assessing the Interpretability of Programmatic Policies with Large Language Models
Zahra Bashir, Michael Bowling, Levi H. S. Lelis
Language Model Inherent Interpretability Programmatic Policy

November 8, 2023

Interpreting Pretrained Language Models via Concept Bottlenecks
Zhen Tan, Lu Cheng, Song Wang, Yuan Bo, Jundong Li, Huan Liu
Language Model Inherent Interpretability Label Noise Pretrained Language Model Concept Bottleneck Model Attention Weight

November 4, 2023

A Survey of the Various Methodologies Towards making Artificial Intelligence More Explainable
Sopam Dasgupta
Artificial Intelligence Timely Survey Inherent Interpretability Complex Reasoning Line by Line Explanation High Explainability Practical Method

October 31, 2023

Interpretable Neural PDE Solvers using Symbolic Frameworks
Yolanne Yi Ran Lee
Inherent Interpretability Symbolic Regression Neural Solver Neural PDE Solver Convolutional Operation

October 30, 2023

October 25, 2023

On the Identifiability and Interpretability of Gaussian Process Models
Jiawen Chen, Wancen Mu, Yun Li, Didong Li
Inherent Interpretability Gaussian Process Mixture Model Novel Identifiability Test Mat\'ern Kernel

Inherent Interpretability

Papers

FeaInfNet: Diagnosis in Medical Image with Feature-Driven Inference and Visual Explanations

AS-XAI: Self-supervised Automatic Semantic Interpretation for CNN

Benchmarking and Enhancing Disentanglement in Concept-Residual Models

CLIP-QDA: An Explainable Concept Bottleneck Model

A data-science pipeline to enable the Interpretability of Many-Objective Feature Selection

Interpreting Differentiable Latent States for Healthcare Time-series Data

Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patching

XAI for time-series classification leveraging image highlight methods

Everybody Needs a Little HELP: Explaining Graphs via Hierarchical Concepts

On the Relationship Between Interpretability and Explainability in Machine Learning

An NMF-Based Building Block for Interpretable Neural Networks With Continual Learning

Rethinking Large Language Models in Mental Health Applications

Finding and Editing Multi-Modal Neurons in Pre-Trained Transformers

Assessing the Interpretability of Programmatic Policies with Large Language Models

Interpreting Pretrained Language Models via Concept Bottlenecks

A Survey of the Various Methodologies Towards making Artificial Intelligence More Explainable

Interpretable Neural PDE Solvers using Symbolic Frameworks

Faithful and Robust Local Interpretability for Textual Predictions

Interpretable-by-Design Text Understanding with Iteratively Generated Concept Bottleneck

On the Identifiability and Interpretability of Gaussian Process Models