ML @ CMU

MARCH 31, 2023

This blog post discusses the effectiveness of black-box model explanations in aiding end users to make decisions. Our work further motivates novel directions for developing and evaluating tools to support human-ML interactions. How can we better support human-ML interactions?

ML 98

ML Explainability Algorithm Machine Learning 98

ML @ CMU

MARCH 30, 2023

This blog post discusses the effectiveness of black-box model explanations in aiding end users to make decisions. Our work further motivates novel directions for developing and evaluating tools to support human-ML interactions. How can we better support human-ML interactions?

ML 52

ML Explainability Algorithm Machine Learning 52

Heartbeat

SEPTEMBER 13, 2023

Source: ResearchGate Explainability refers to the ability to understand and evaluate the decisions and reasoning underlying the predictions from AI models (Castillo, 2021). Cynthia Rudin, a computer science professor at Duke University, emphasized the difference between interpretability and explainability. Russell, C. &

Explainability 52

Explainability Machine Learning Deep Learning Explainable AI 52

Mlearning.ai

FEBRUARY 6, 2023

These issues are localized to OpenAI’s Text Classifier specifically and may not generalize to production-ready AI-Detectors in general. There are plenty of techniques to help reduce overfitting in ML models. For example, Stanford received around 55,471 applications in 2021 [5]. This is why we need Explainable AI (XAI).

Explainable AI 52

Explainable AI Explainability ChatGPT Chatbots 52

The MLOps Blog

AUGUST 11, 2023

There comes a time when every ML practitioner realizes that training a model in Jupyter Notebook is just one small part of the entire project. At that point, the Data Scientists or ML Engineers become curious and start looking for such implementations. What are ML pipeline architecture design patterns?

ML 52

ML Machine Learning Data Ingestion Deep Learning 52

Mlearning.ai

FEBRUARY 7, 2023

Some of the posts, blogs, and articles dealing with this new phenomenom, well, really don’t deserve any attention. 1x: A nice prompt forcing the AI to interrupt itself while explaining AI alignment. ChatGPT is the first true foundation model for the mass-market. But some of them are genuinely interesting.

ChatGPT 105

ChatGPT Large Language Models Explainability Machine Learning 105