Pseudo Stochastic

Pseudo-stochasticity refers to systems exhibiting deterministic behavior conditional on underlying stochastic processes, a concept finding applications in diverse fields like data assimilation, reinforcement learning, and natural language processing. Current research focuses on developing efficient algorithms to handle these systems, including methods leveraging denoising diffusion models for ensemble generation, augmented data sampling for improved reinforcement learning, and novel recurrent neural network architectures like PVGRU for enhanced dialogue generation. These advancements aim to improve the accuracy and efficiency of models dealing with complex, partially deterministic systems, ultimately impacting fields ranging from weather forecasting to chatbot development.

Papers

November 4, 2024

Pseudo-Probability Unlearning: Towards Efficient and Privacy-Preserving Machine Unlearning
Zihao Zhao, Yijiang Li, Yuchen Yang, Wenqing Zhang, Nuno Vasconcelos, Yinzhi Cao
Membership Inference Attack Privacy Preserving Anti Forgetting Machine Unlearning Pseudo Stochastic

August 13, 2023

Generating observation guided ensembles for data assimilation with denoising diffusion probabilistic model
Yuuichi Asahi, Yuta Hasegawa, Naoyuki Onodera, Takashi Shimokawabe, Hayato Shiba, Yasuhiro Idomura
Diffusion Explainer Probabilistic Model Diverse Ensemble Data Assimilation Generating Future Observation Pseudo Stochastic

May 25, 2023

Sample Efficient Reinforcement Learning in Mixed Systems through Augmented Samples and Its Applications to Queueing Networks
Honghao Wei, Xin Liu, Weina Wang, Lei Ying
Financial Application Network Programming Deep Q Learning Efficient Reinforcement Learning Sample Efficient Reinforcement Learning Hybrid System Augmented Sample Pseudo Stochastic

December 18, 2022

PVGRU: Generating Diverse and Relevant Dialogue Responses via Pseudo-Variational Mechanism
Yongkang Liu, Shi Feng, Daling Wang, Yifei Zhang, Hinrich Schütze
Generative AI Recurrent Neural Network Response Generation Diverse Set Gated Recurrent Unit Dialog Model Dialogue Response Pseudo Stochastic

February 22, 2022

Nonconvex Extension of Generalized Huber Loss for Robust Learning and Pseudo-Mode Statistics
Kaan Gokcesu, Hakan Gokcesu
Robust Learning Convex Loss General Nonconvex Robust Loss Function Huber Loss Exponential Convergence Pseudo Stochastic

Pseudo Stochastic

Papers

Pseudo-Probability Unlearning: Towards Efficient and Privacy-Preserving Machine Unlearning

Generating observation guided ensembles for data assimilation with denoising diffusion probabilistic model

Sample Efficient Reinforcement Learning in Mixed Systems through Augmented Samples and Its Applications to Queueing Networks

PVGRU: Generating Diverse and Relevant Dialogue Responses via Pseudo-Variational Mechanism

Nonconvex Extension of Generalized Huber Loss for Robust Learning and Pseudo-Mode Statistics