What is recorded here?
Proposed sensitivity and implementation-invariance axioms for feature attribution and derived the integrated-gradients method, then illustrated it on image, text and chemistry models.
Source-linked catalogue note. A detailed explanatory guide has not yet been prepared for this entry.
Scientific scope
The record is the exact three-author ICML 2017 paper. Attributions are model- and baseline-dependent explanations of predictions, not causal effects, guarantees of faithfulness in every setting or evidence that a model is safe.
Read the original work
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Mukund Sundararajan; Ankur Taly; Qiqi Yan
The named authors contributed to this scoped work; individual effort shares and later applications are not assigned by the byline.
Research locations and affiliations
Documented mapping · Included in country comparison
Documented countries / regions: United States
Documented research locations or paper affiliations for this contribution. Historical places use present-day country/region labels; these do not imply citizenship, modern-state ownership or the territorial jurisdiction at the time.
English catalogue narrative
Sundararajan, Taly and Yan: integrated gradients
Proposed sensitivity and implementation-invariance axioms for feature attribution and derived the integrated-gradients method, then illustrated it on image, text and chemistry models.
Scientific scope
The record is the exact three-author ICML 2017 paper. Attributions are model- and baseline-dependent explanations of predictions, not causal effects, guarantees of faithfulness in every setting or evidence that a model is safe.
Reading status
This is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.
People and systems
Mukund Sundararajan; Ankur Taly; Qiqi Yan
The named authors contributed to this scoped work; individual effort shares and later applications are not assigned by the byline.