yoavg

Reinforcement Learning for Language Models

Yoav Goldberg, April 2023.

Why RL?

With the release of the ChatGPT model and followup large language models (LLMs), there was a lot of discussion of the importance of "RLHF training", that is, "reinforcement learning from human feedback". I was puzzled for a while as to why RL (Reinforcement Learning) is better than learning from demonstrations (a.k.a supervised learning) for training language models. Shouldn't learning from demonstrations (or, in language model terminology "instruction fine tuning", learning to immitate human written answers) be sufficient? I came up with a theoretical argument that was somewhat convincing. But I came to realize there is an additional argumment which not only supports the case of RL training, but also requires it, in particular for models like ChatGPT. This additional argument is spelled out in (the first half of) a talk by John Schulman from OpenAI. This post pretty much

mbarretta

ECK storage recommendations

Table of contents

• tl;dr (short version)
• Kubernetes storage options
  • emptyDir
  • hostPath
  • PersistentVolumes
• PersistentVolumes in details

loretoparisi

#!/usr/local/bin/python3
# @author cpuhrsch https://github.com/cpuhrsch
# @author Loreto Parisi loreto@musixmatch.com

import argparse
import numpy as np
from sklearn.metrics import confusion_matrix

def parse_labels(path):
    with open(path, 'r') as f:

iros

Title

<Additional information about your API call. Try to use verbs that match both request type (fetching vs modifying) and plurality (one vs multiple).>

• URL

  <The URL Structure (path only, no root url)>

• Method: