SOLVED: A potential limitation of relying heavily on human feedback in RLHF, particularl ...

A potential limitation of relying heavily on human feedback in RLHF, particularly when dealing with subjective or creative tasks is that

Human feedback is always perfectly objective and unbiased

Human preferences can be difficult to quantify and translate into a robust reward signal

Human feedback is readily available in unlimited quantities and at no cost

Human feedback is guaranteed to align perfectly with the long term goals of the AI system

Verified Answer

Correct Option - b

To get all Infosys Certified Generative AI Professional Advanced Exam questions Join Telegram Group https://rebrand.ly/lex-group-70b557

We're passionate about offering best placement materials and courses!! A one stop place for Placement Materials. We daily post Offcampus updates and Placement Materials.

Important Links

Privacy Policy

Refund Policy

Terms & Conditions

About Us

Contact Us