Author name cluster

Yu Duan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

AAAI Conference 2026 Conference Paper

Learning Intrinsic Hierarchy for Generalized Category Discovery

Yu Duan
Junzhi He
Zhanxuan Hu
Mengda Ji
Rong Wang
Quanxue Gao

Generalized Category Discovery (GCD) aims to classify unlabeled data by leveraging knowledge from labeled categories. While existing methods have achieved remarkable progress, they often treat images as flat feature sets, neglecting the intrinsic hierarchy: where key objects dominate meaning and backgrounds serve as context. For instance, in images of a dog either standing on grass or lying on a bed, the dog remains the central semantic element, whereas the background varies. Motivated by this, we propose LEArning Intrinsic Hierarchy (LEAH), a lightweight plug-and-play module designed to model hierarchical structure within images. LEAH consists of two components: a pruner that filters task-irrelevant tokens to extract key objects, and a constructor that embeds key objects and full images into hyperbolic space using adaptive entailment cones to capture compositional semantics. LEAH can be easily integrated into existing GCD frameworks with minimal modification. When applied to SimGCD, it achieves up to 13.2% accuracy improvement on fine-grained benchmarks, demonstrating its effectiveness in discovering subtle inter-class differences through hierarchical modeling.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Maximizing Schatten-p Norm Regularization Toward Balance

Fangfang Li
Quanxue Gao
Yapeng Wang
Yu Duan
Yuzhuo Feng
Qin Li

The Schatten-p norm, as a class of structure-inducing norms based on singular values, has been widely used to enhance model low-rankness and representation capability due to its flexibility in structural modeling and favorable mathematical properties. However, its potential in cluster distribution modeling has long been overlooked. Therefore, we explore the potential of maximizing the Schatten-p norm as a regularization strategy specifically designed to achieve balanced clustering. This work is the first to investigate its effectiveness in promoting cluster balance. To be specific, maximizing Schatten-p norm effectively guides the assignment of data points, ensuring a more balanced distribution of samples across clusters. We have conducted an in-depth theoretical analysis and validated its effectiveness through extensive clustering experiments. Experimental results demonstrate that, compared to existing methods, this regularization term significantly improves clustering quality and obtain reasonable clustering.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

POCO: Scalable Neural Forecasting through Population Conditioning

Yu Duan
Hamza Chaudhry
Misha B Ahrens
Christopher Harvey
Matthew G Perich
Karl Deisseroth
Kanaka Rajan

Predicting future neural activity is a core challenge in modeling brain dynamics, with applications ranging from scientific investigation to closed-loop neurotechnology. While recent models of population activity emphasize interpretability and behavioral decoding, neural forecasting—particularly across multi-session, spontaneous recordings—remains underexplored. We introduce POCO, a unified forecasting model that combines a lightweight univariate forecaster with a population-level encoder to capture both neuron-specific and brain-wide dynamics. Trained across five calcium imaging datasets spanning zebrafish, mice, and C. elegans, POCO achieves state-of-the-art accuracy at cellular resolution in spontaneous behaviors. After pre-training, POCO rapidly adapts to new recordings with minimal fine-tuning. Notably, POCO's learned unit embeddings recover biologically meaningful structure—such as brain region clustering—without any anatomical labels. Our comprehensive analysis reveals several key factors influencing performance, including context length, session diversity, and preprocessing. Together, these results position POCO as a scalable and adaptable approach for cross-session neural forecasting and offer actionable insights for future model design. By enabling accurate, generalizable forecasting models of neural dynamics across individuals and species, POCO lays the groundwork for adaptive neurotechnologies and large-scale efforts for neural foundation models. Code is available at https: //github. com/yuvenduan/POCO.

PDF Details

EAAI Journal 2024 Journal Article

Adversarial deep reinforcement learning based robust depth tracking control for underactuated autonomous underwater vehicle

Zhao Wang
Xianbo Xiang
Yu Duan
Shaolong Yang

In this paper, an adversarial deep reinforcement learning-based control method is proposed to address the issue of robust depth tracking of an underactuated autonomous underwater vehicle in the presence of intrinsic coupled dynamics and external disturbances. First, long-short-term-memory neural network is presented to memorize and predict the changes in the state of vehicle, and a cascaded multilayer perception projects the output into action space of vehicle. Subsequently, adversarial deep reinforcement learning scheme is applied to the training of control agent by introducing an adversary which counteracts the control behavior, whereby the agent is enabled to learn the control strategy in different distributions of state transition. For evaluation of the performance, a control agent is pre-trained in simulation environment based on the reliable digital model of a real vehicle, and the simulation environment is paced at one iteration per second to align with real-time operations to ensure the portability of training result. Furthermore, the training cost is also extremely reduced. Finally, experiments are conducted with time-varying disturbances to further prove the feasibility of the proposed learning-based control scheme on a prototype of underwater vehicle in towing tank. Moreover, comparative experiment results show the better robustness performance of the learning-based control agent than that of classic line-of-sight based proportional–integral–derivative and adaptive line-of-sight based proportional–integral–derivative controllers in different scenarios.

Details DOI

ICLR Conference 2023 Conference Paper

Hebbian and Gradient-based Plasticity Enables Robust Memory and Rapid Learning in RNNs

Yu Duan
Zhongfan Jia
Qian Li
Yi Zhong
Kaisheng Ma

Rapidly learning from ongoing experiences and remembering past events with a flexible memory system are two core capacities of biological intelligence. While the underlying neural mechanisms are not fully understood, various evidence supports that synaptic plasticity plays a critical role in memory formation and fast learning. Inspired by these results, we equip Recurrent Neural Networks (RNNs) with plasticity rules to enable them to adapt their parameters according to ongoing experiences. In addition to the traditional local Hebbian plasticity, we propose a global, gradient-based plasticity rule, which allows the model to evolve towards its self-determined target. Our models show promising results on sequential and associative memory tasks, illustrating their ability to robustly form and retain memories. In the meantime, these models can cope with many challenging few-shot learning problems. Comparing different plasticity rules under the same framework shows that Hebbian plasticity is well-suited for several memory and associative learning tasks; however, it is outperformed by gradient-based plasticity on few-shot regression tasks which require the model to infer the underlying mapping.

Details

IROS Conference 2018 Conference Paper

Optimal Feedback Control Based on Analytical Linear Models Extracted from Neural Networks Trained for Nonlinear Systems

Yu Duan
Shuhei Ikemoto
Koh Hosoda

A number of researches have been focusing on the development and control of robots with soft structures such as flexible musculoskeletal systems. Thus far, it has been reported that these robots can achieve high adaptability to environments despite their extremely simple controllers. However, because these robots are difficult to model mathematically, there is still no systematic design policy, in which control theory has been playing a role in conventional robotics, for constituting simple controllers. To tackle this problem, we propose a new approach using a neural network to obtain mathematical models. In particular, with this method, the control theory is applied to linear system models extracted from a network trained to express the forward dynamics of a robot. Through simulations, the validity and advantage of the proposed method was successfully confirmed.

Details