Implement an inverse-propensity-weighted (IPS) off-policy evaluator for logged agent actio

Hard

Agents, Tool Use & Product Post-Training6 / 7

Chapter 196 / 7

Agents, Tool Use & Product Post-Training

Implement an inverse-propensity-weighted (IPS) off-policy evaluator for logged agent actio

Hard

Implement an inverse-propensity-weighted (IPS) off-policy evaluator for logged agent actions.

Target interface

ips_evaluate(logs, target_policy, clip=None, self_normalize=True)

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.

solution.pyPython

Local draft

import numpy as np

def ips_evaluate(logs, target_policy, clip=None, self_normalize=True):
    raise NotImplementedError

⌘/Ctrl↵to submit

AI review

No review yet

Submit your solution for a verdict, a score out of 100 and concrete feedback. Any correct approach passes.

Chapter 196 / 7

Agents, Tool Use & Product Post-Training

Implement an inverse-propensity-weighted (IPS) off-policy evaluator for logged agent actio

Hard

Implement an inverse-propensity-weighted (IPS) off-policy evaluator for logged agent actions.

Target interface

ips_evaluate(logs, target_policy, clip=None, self_normalize=True)

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.