Zhaohui Jiang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhaohui Jiang has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Paul Weng
Duke Kunshan University, Kunshan, China
5 shared publications Xuening Feng
Shanghai Jiao Tong University
3 shared publications- Yifei Zhu
Global College, Shanghai Jiao Tong University, Shanghai, China
2 shared publications Changjie Fan
Electrical and Computer Engineering AI Lab The University of Michigan-Shanghai Jiao Tong University Joint Institute (UM-SJTU JI) Duke Kunshan University NetEase Fuxi
1 shared publicationClaire Glanois
1 shared publication- Dong Li
Huawei Noah's Ark Lab
1 shared publication
Latest Publications
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12045-12053
Progressive Learning with Human Feedback for Personalized Adaptive Video Streaming
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12045-12053
Open AccessProgressive Learning with Human Feedback for Personalized Adaptive Video Streaming
- Open Access
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (16) , 16604-16612
DUO: Diverse, Uncertain, On-Policy Query Generation and Selection for Reinforcement Learning from Human Feedback
Defining a reward function is usually a challenging but critical task for the system designer in reinforcement learning, especially when specifying complex behaviors. Reinforcement learning from human feedback (RLHF) emerges as a promising approach to circumvent this. In RLHF, the agent typically learns a reward function by querying a human teacher using pairwise comparisons of trajectory segments. A key question in this domain is how to reduce the number of queries necessary to learn an informative reward function since asking a human teacher too many queries is impractical and costly. To tackle this question, we propose DUO, a novel method for diverse, uncertain, on-policy query generation and selection in RLHF. Our method produces queries that are (1) more relevant for policy training (via an on-policy criterion), (2) more informative (via a principled measure of epistemic uncertainty), and (3) diverse (via a clustering-based filter). Experimental results on a variety of locomotion and robotic manipulation tasks demonstrate that our method can outperform state-of-the-art RLHF methods given the same total budget of queries, while being robust to possibly irrational teachers.
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (16) , 16604-16612
Open AccessDUO: Diverse, Uncertain, On-Policy Query Generation and Selection for Reinforcement Learning from Human Feedback
Defining a reward function is usually a challenging but critical task for the system designer in reinforcement learning, especially when specifying complex behaviors. Reinforcement learning from human feedback (RLHF) emerges as a promising approach to circumvent this. In RLHF, the agent typically learns a reward function by querying a human teacher using pairwise comparisons of trajectory segments. A key question in this domain is how to reduce the number of queries necessary to learn an informative reward function since asking a human teacher too many queries is impractical and costly. To tackle this question, we propose DUO, a novel method for diverse, uncertain, on-policy query generation and selection in RLHF. Our method produces queries that are (1) more relevant for policy training (via an on-policy criterion), (2) more informative (via a principled measure of epistemic uncertainty), and (3) diverse (via a clustering-based filter). Experimental results on a variety of locomotion and robotic manipulation tasks demonstrate that our method can outperform state-of-the-art RLHF methods given the same total budget of queries, while being robust to possibly irrational teachers.
- Open Access
Reinforcement Learning From Imperfect Corrective Actions And Proxy Rewards
In practice, reinforcement learning (RL) agents are often trained with a possibly imperfect proxy reward function, which may lead to a human-agent alignment issue (i.e., the learned policy either converges to non-optimal performance with low cumulative rewards, or achieves high cumulative rewards but in undesired manner). To tackle this issue, we consider a framework where a human labeler can provide additional feedback in the form of corrective actions, which expresses the labeler's action preferences although this feedback may possibly be imperfect as well. In this setting, to obtain a better-aligned policy guided by both learning signals, we propose a novel value-based deep RL algorithm called Iterative learning from Corrective actions and Proxy rewards (ICoPro), which cycles through three phases: (1) Solicit sparse corrective actions from a human labeler on the agent's demonstrated trajectories; (2) Incorporate these corrective actions into the Q-function using a margin loss to enforce adherence to labeler's preferences; (3) Train the agent with standard RL losses regularized with a margin loss to learn from proxy rewards and propagate the Q-values learned from human feedback. Moreover, another novel design in our approach is to integrate pseudo-labels from the target Q-network to reduce human labor and further stabilize training. We experimentally validate our proposition on a variety of tasks (Atari games and autonomous driving on highway). On the one hand, using proxy rewards with different levels of imperfection, our method can better align with human preferences and is more sample-efficient than baseline methods. On the other hand, facing corrective actions with different types of imperfection, our method can overcome the non-optimality of this feedback thanks to the guidance from proxy reward.
Open AccessReinforcement Learning From Imperfect Corrective Actions And Proxy Rewards
In practice, reinforcement learning (RL) agents are often trained with a possibly imperfect proxy reward function, which may lead to a human-agent alignment issue (i.e., the learned policy either converges to non-optimal performance with low cumulative rewards, or achieves high cumulative rewards but in undesired manner). To tackle this issue, we consider a framework where a human labeler can provide additional feedback in the form of corrective actions, which expresses the labeler's action preferences although this feedback may possibly be imperfect as well. In this setting, to obtain a better-aligned policy guided by both learning signals, we propose a novel value-based deep RL algorithm called Iterative learning from Corrective actions and Proxy rewards (ICoPro), which cycles through three phases: (1) Solicit sparse corrective actions from a human labeler on the agent's demonstrated trajectories; (2) Incorporate these corrective actions into the Q-function using a margin loss to enforce adherence to labeler's preferences; (3) Train the agent with standard RL losses regularized with a margin loss to learn from proxy rewards and propagate the Q-values learned from human feedback. Moreover, another novel design in our approach is to integrate pseudo-labels from the target Q-network to reduce human labor and further stabilize training. We experimentally validate our proposition on a variety of tasks (Atari games and autonomous driving on highway). On the one hand, using proxy rewards with different levels of imperfection, our method can better align with human preferences and is more sample-efficient than baseline methods. On the other hand, facing corrective actions with different types of imperfection, our method can overcome the non-optimality of this feedback thanks to the guidance from proxy reward.
- Open Access
Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning
To improve the sample efficiency of vision-based deep reinforcement learning (RL), we propose a novel method, called SPIRL, to automatically extract important patches from input images. Following Masked Auto-Encoders, SPIRL is based on Vision Transformer models pre-trained in a self-supervised fashion to reconstruct images from randomly-sampled patches. These pre-trained models can then be exploited to detect and select salient patches, defined as hard to reconstruct from neighboring patches. In RL, the SPIRL agent processes selected salient patches via an attention module. We empirically validate SPIRL on Atari games to test its data-efficiency against relevant state-of-the-art methods, including some traditional model-based methods and keypoint-based models. In addition, we analyze our model's interpretability capabilities.
Open AccessUnsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning
To improve the sample efficiency of vision-based deep reinforcement learning (RL), we propose a novel method, called SPIRL, to automatically extract important patches from input images. Following Masked Auto-Encoders, SPIRL is based on Vision Transformer models pre-trained in a self-supervised fashion to reconstruct images from randomly-sampled patches. These pre-trained models can then be exploited to detect and select salient patches, defined as hard to reconstruct from neighboring patches. In RL, the SPIRL agent processes selected salient patches via an attention module. We empirically validate SPIRL on Atari games to test its data-efficiency against relevant state-of-the-art methods, including some traditional model-based methods and keypoint-based models. In addition, we analyze our model's interpretability capabilities.
- 18 September 2023
- Published by Springer Nature
- p. 556-572
Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning
- 18 September 2023
- Published by Springer Nature
- p. 556-572
Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning
- Open Access
Differentiable Logic Machines
The integration of reasoning, learning, and decision-making is key to build more general artificial intelligence systems. As a step in this direction, we propose a novel neural-logic architecture, called differentiable logic machine (DLM), that can solve both inductive logic programming (ILP) and reinforcement learning (RL) problems, where the solution can be interpreted as a first-order logic program. Our proposition includes several innovations. Firstly, our architecture defines a restricted but expressive continuous relaxation of the space of first-order logic programs by assigning weights to predicates instead of rules, in contrast to most previous neural-logic approaches. Secondly, with this differentiable architecture, we propose several (supervised and RL) training procedures, based on gradient descent, which can recover a fully-interpretable solution (i.e., logic formula). Thirdly, to accelerate RL training, we also design a novel critic architecture that enables actor-critic algorithms. Fourthly, to solve hard problems, we propose an incremental training procedure that can learn a logic program progressively. Compared to state-of-the-art (SOTA) differentiable ILP methods, DLM successfully solves all the considered ILP problems with a higher percentage of successful seeds (up to 3.5$\times$). On RL problems, without requiring an interpretable solution, DLM outperforms other non-interpretable neural-logic RL approaches in terms of rewards (up to 3.9%). When enforcing interpretability, DLM can solve harder RL problems (e.g., Sorting, Path) Moreover, we show that deep logic programs can be learned via incremental supervised training. In addition to this excellent performance, DLM can scale well in terms of memory and computational time, especially during the testing phase where it can deal with much more constants ($>$2$\times$) than SOTA.
Open AccessDifferentiable Logic Machines
The integration of reasoning, learning, and decision-making is key to build more general artificial intelligence systems. As a step in this direction, we propose a novel neural-logic architecture, called differentiable logic machine (DLM), that can solve both inductive logic programming (ILP) and reinforcement learning (RL) problems, where the solution can be interpreted as a first-order logic program. Our proposition includes several innovations. Firstly, our architecture defines a restricted but expressive continuous relaxation of the space of first-order logic programs by assigning weights to predicates instead of rules, in contrast to most previous neural-logic approaches. Secondly, with this differentiable architecture, we propose several (supervised and RL) training procedures, based on gradient descent, which can recover a fully-interpretable solution (i.e., logic formula). Thirdly, to accelerate RL training, we also design a novel critic architecture that enables actor-critic algorithms. Fourthly, to solve hard problems, we propose an incremental training procedure that can learn a logic program progressively. Compared to state-of-the-art (SOTA) differentiable ILP methods, DLM successfully solves all the considered ILP problems with a higher percentage of successful seeds (up to 3.5$\times$). On RL problems, without requiring an interpretable solution, DLM outperforms other non-interpretable neural-logic RL approaches in terms of rewards (up to 3.9%). When enforcing interpretability, DLM can solve harder RL problems (e.g., Sorting, Path) Moreover, we show that deep logic programs can be learned via incremental supervised training. In addition to this excellent performance, DLM can scale well in terms of memory and computational time, especially during the testing phase where it can deal with much more constants ($>$2$\times$) than SOTA.