H Choi, S Park - IEEE Access, 2022 - ieeexplore.ieee.org
… a cachereplacement policy technique to increase the cache hit rate. This policy can improve the efficiency of cache … Ahn, “An imitationlearningapproach for cachereplacement,” in In: …
S Qiu, Q Fan, X Li, X Zhang, G Min… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
… We propose an imitationlearning architecture to calculate the eviction probability distribution to approximate the oracle policy. We innovatively introduce the binary crossentropy loss …
… imitationlearning to train cachereplacement policies to mimic an optimal replacement … this work will focus on simple caches that are effective for evaluating cachereplacement policies. …
Y Zhou, F Wang, Z Shi, D Feng - Proceedings of the International …, 2022 - ojs.aaai.org
… the utility of deep reinforcement learning (DRL) in cachereplacement policies. Most of the … block in the cache and uses heuristic or ML methods to design cachereplacement policies. In …
Z Yan, P Cheng, Z Chen, B Vucetic… - IEEE/ACM Transactions …, 2021 - ieeexplore.ieee.org
… In this paper, we propose a novel Gaussian process imitationlearning (GPIL) method to … GPIL method lies in exploiting the unique enumeration tree structure of B&B, and learning how …
… , this paper proposes an imitationlearning enabled task … it does not require RSUs to cache or process computational tasks for … , we design an imitationlearning-based method to conquer …
… into other research fields like scheduling [35] and cachereplacement problem [25]. … [46] proposed thirdperson imitationlearning (TPIL) which partition the discriminator into feature …
X Wang, Z Ning, S Guo - IEEE Transactions on Parallel and …, 2020 - ieeexplore.ieee.org
… a multiagent imitationlearning based computation offloading algo… computing based on multiagent imitationlearning. Generally, … , communication and caching scheduling among edge …
Y Hu, M Wang, Y Chen, C Fan - 2021 IEEE Conference on …, 2021 - ieeexplore.ieee.org
… imitationlearning based (IL-based) method and reinforcement learning based (RL-based) method to infer the data that are most likely to be used in the future by learning the behaviour …