SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.10342 · ICLR · 2023

Fantastic Rewards and How to Tame Them: A Case Study on Reward Learning for Task-oriented Dialogue Systems

Mingyuan Zhou, Huan Wang, Caiming Xiong, Yihao Feng, Shujian Zhang, Shentao Yang, Jianguo Zhang

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 11 functions out of this paper's own repositories and ran 7 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
shentao-yang/fantastic_reward_iclr2023 canonical 7 of 11
FunctionStatusWhere it lives
binaryTupleList2Tensor Ran shentao-yang/fantastic_reward_iclr2023/RewardTorch.py
code served (permissive licence) · get_code("c1f35d36da13f065")
cuda_ Ran shentao-yang/fantastic_reward_iclr2023/BART.py
code served (permissive licence) · get_code("b849278a0ba6c25c")
get_state Ran shentao-yang/fantastic_reward_iclr2023/EstimateBehaviorPolicy.py
code served (permissive licence) · get_code("7f76b00aa1bb20bf")
get_turn_state Ran shentao-yang/fantastic_reward_iclr2023/EstimateBehaviorPolicy.py
code served (permissive licence) · get_code("9f2a9615de6e84bd")
invert_mask Ran shentao-yang/fantastic_reward_iclr2023/BART.py
code served (permissive licence) · get_code("c9f6adbecd206460")
puntuation_handler Ran shentao-yang/fantastic_reward_iclr2023/utils.py
code served (permissive licence) · get_code("a9b8c7a10994073f")
shift_tokens_right Ran shentao-yang/fantastic_reward_iclr2023/BART.py
code served (permissive licence) · get_code("e00ccf7847d12b59")
clean_time Not yet run shentao-yang/fantastic_reward_iclr2023/damd_multiwoz/clean_dataset.py
code served (permissive licence) · get_code("4c62f23ce8135bda")
get_state_act Not yet run shentao-yang/fantastic_reward_iclr2023/EstimateBehaviorPolicy.py
code served (permissive licence) · get_code("a3f6d589f208358b")
numpy2torch Not yet run shentao-yang/fantastic_reward_iclr2023/RewardTorch.py
code served (permissive licence) · get_code("54302690d7f170c5")
numpyBinaryTuple2torch Not yet run shentao-yang/fantastic_reward_iclr2023/RewardTorch.py
code served (permissive licence) · get_code("ef7fddf0c4145890")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

When learning task-oriented dialogue (ToD) agents, reinforcement learning (RL) techniques can naturally be utilized to train dialogue strategies to achieve userspecific goals. Prior works mainly focus on adopting advanced RL techniques to train the ToD agents, while the design of the reward function is not well studied. This paper aims at answering the question of how to efficiently learn and leverage a reward function for training end-to-end (E2E) ToD agents. Specifically, we introduce two generalized objectives for reward-function learning, inspired by the classical learning-to-rank literature. Further, we utilize the learned reward function to guide the training of the E2E ToD agent. With the proposed techniques, we achieve competitive results on the E2E response-generation task on the Multiwoz 2.0 dataset. Source code and checkpoints are publicly released at https://github.com/Shentao-YANG/Fantastic Reward ICLR2023.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.10342")
get_code_for_paper("2302.10342")
have("2302.10342")

Connect an agent — have() is free.