#contrastiveexplanations Suchergebnisse

News from the #explainability front-line in our lab meeting: André Artelt introduced #ContrastiveExplanations for #MachineLearning and suggested an efficient way to compute them, based on his previous work on #CounterfactualExplanations.

HammerLabML's tweet image. News from the #explainability front-line in our lab meeting: André Artelt introduced #ContrastiveExplanations for #MachineLearning and suggested an efficient way to compute them, based on his previous work on #CounterfactualExplanations.

💥Conference news 💥: Our work "Efficient computation of #ContrastiveExplanations" by André Artelt and Barbara Hammer got accepted at #IJCNN2021. Preprint on arXiv: arxiv.org/pdf/2010.02647…


#Contrastiveexplanations: the question “why did Liz open the door” has many possible contrasts that provides hints-> “why did Liz open the door ...instead of leaving it closed... instead of the window... instead of Michael opening it”. Each calls for a different explanation 🙌


#contrastiveexplanations: research indicates that people request only contrastive explanations, and that the cognitive burden of the complete explanation is too great. We are wired this way -- can argue that a layperson will find contrastive explanations more intuitive & valuable


New paper alert🥳 A preprint of our work on fairness and robustness of #ContrastiveExplanations by André Artelt and Barbara Hammer is available on arXiv: arxiv.org/pdf/2103.02354… #MachineLearning #ExplainableAI


💥Conference news 💥: Our work "Efficient computation of #ContrastiveExplanations" by André Artelt and Barbara Hammer got accepted at #IJCNN2021. Preprint on arXiv: arxiv.org/pdf/2010.02647…


New paper alert🥳 A preprint of our work on fairness and robustness of #ContrastiveExplanations by André Artelt and Barbara Hammer is available on arXiv: arxiv.org/pdf/2103.02354… #MachineLearning #ExplainableAI


Revised version of our paper on the efficient computation of #ContrastiveExplanations by André Artelt and Barbara Hammer is now available on #arXiv #machinelearning - check it out: arxiv.org/abs/2010.02647


News from the #explainability front-line in our lab meeting: André Artelt introduced #ContrastiveExplanations for #MachineLearning and suggested an efficient way to compute them, based on his previous work on #CounterfactualExplanations.

HammerLabML's tweet image. News from the #explainability front-line in our lab meeting: André Artelt introduced #ContrastiveExplanations for #MachineLearning and suggested an efficient way to compute them, based on his previous work on #CounterfactualExplanations.

#contrastiveexplanations: research indicates that people request only contrastive explanations, and that the cognitive burden of the complete explanation is too great. We are wired this way -- can argue that a layperson will find contrastive explanations more intuitive & valuable


#Contrastiveexplanations: the question “why did Liz open the door” has many possible contrasts that provides hints-> “why did Liz open the door ...instead of leaving it closed... instead of the window... instead of Michael opening it”. Each calls for a different explanation 🙌


Keine Ergebnisse für "#contrastiveexplanations"

News from the #explainability front-line in our lab meeting: André Artelt introduced #ContrastiveExplanations for #MachineLearning and suggested an efficient way to compute them, based on his previous work on #CounterfactualExplanations.

HammerLabML's tweet image. News from the #explainability front-line in our lab meeting: André Artelt introduced #ContrastiveExplanations for #MachineLearning and suggested an efficient way to compute them, based on his previous work on #CounterfactualExplanations.

Loading...

Something went wrong.


Something went wrong.


United States Trends