SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.08968 · CVPR · 2024

MCPNet: An Interpretable Classifier via Multi-Level Concept Prototypes

Wei-Chen Chiu, Chien-Yi Wang, Bor-Shiun Wang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Concept Prototype #B (from Layer 2) Concept Prototype #A (from Layer 1) Concept Prototype #D (from Layer 4) Concept Prototype #C (from Layer 3) Concept Strength Bobcat Concept Strength Fox Concept Strength PIP-Net MCPNet Label : Bobcat Image MCP distribution e.g. Brown color e.g. White color e.g. Bobcat e.g. Dapple Figure 1. The "Bobcat" was correctly classified by our MCPNet but incorrectly classified as a "Fox" by PIP-Net [15]. On the right side, we provide an illustration of using our proposed Multi-level Concept Prototype (MCP) distribution to classify and explain the input image. In particular, our concept prototypes are extracted from multiple layers of the classification model (thus having low-level to highlevel concepts). In comparison with a recent state-of-the-art baseline, PIP-Net [15] shown on the left side which only adopts single-level explanations (symbolized as colorful boxes on the bottom portion, they are usually extracted from the last model layer), our proposed MCPNet provides more comprehensive explanations as well as better classification performance.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2404.08968")
get_code_for_paper("2404.08968")
have("2404.08968")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2404.08968.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2404.08968.svg)](https://syntology.ai/paper/2404.08968)

Connect an agent — have() is free.