Search

The geometric deformation of viscous fingering is useful for understanding natural multiscale patterns and designing dissipative structures in materials. Although the spatio‐temporal patterns in soft materials are reported previously, there is a lack of research on the spatial finiteness and boundary effects. In this study, the recognition of spatial finiteness in "meniscus splitting phenomena" in aqueous polymer dispersions during water evaporation is demonstrated. By providing heat energy to polymer dispersions in a Hele‐Shaw cell, an interface fluctuation with concentration unevenness is induced to split the evaporative interface. The spatial finiteness of the interface causes asynchronous nucleation, which is demonstrated using polysaccharide dispersions. The results of the quasi‐natural experiments revealed that the nonequilibrium drying/wetting period for repositioning polymer clusters allows for considerable changes in Reynolds number in a low range (<10−6) to form multiple nuclei. This splitting method will be universally useful in various fields, including fluid dynamics, biology, and microfluidics, as well as non‐equilibrium, colloid, interface, polymer, and materials sciences. [ABSTRACT FROM AUTHOR]

The Right to be Forgotten gives a data owner the right to revoke their data from an entity storing it. In the context of federated learning, the Right to be Forgotten requires that, in addition to the data itself, any influence of the data on the FL model must disappear, a process we call “federated unlearning.” The most straightforward and legitimate way to implement federated unlearning is to remove the revoked data and retrain the FL model from scratch. Yet the computational and time overhead associated with fully retraining FL models can be prohibitively expensive. In this article, we take the first step to comprehensively investigate the way to settle the unlearning paradigm in the context of federated learning. First, we define the problem of efficient federated unlearning, including its challenges and goals, and we identify three common types of federated unlearning requests: class unlearning, client unlearning, and sample unlearning. Based on those challenges and goals, a general pipeline is proposed for federated unlearning for the above three types of requests. We revisit how the training data affects the final FL model performance and thereby empowers the proposed framework with the reverse stochastic gradient ascent (SGA) and elastic weight consolidation (EWC). Various experiments are conducted to verify effectiveness of the proposed method in both aspects of unlearning efficacy and efficiency. We believe the proposed method will perform as an essential component for future machine unlearning systems. [ABSTRACT FROM AUTHOR]

As the large-scale deployment of machine learning applications, there is much research attention on exploiting a vast amount of data stored on mobile clients. To preserve data privacy, federated learning has been proposed to enable large-scale machine learning by massive clients without exposing raw data. Existing works of federated learning struggle for accelerating the learning process, but ignore the energy efficiency that is critical for resource-constrained clients. In this article, we propose to improve the energy efficiency of federated learning by lowering CPU cycle frequencies of clients who are faster in the training group. Based on this idea, we formulate an optimization problem aiming to minimize the total system cost defined as a weighted sum of learning time and energy consumption. Due to the hardness of the formulated optimization problem and unpredictability of network quality, we propose L4L (Learning for Learning), an experience-driven computational resource control approach based on the deep reinforcement learning, which can derive the near-optimal solution with only the clients’ bandwidth information in the previous training rounds. We conduct the experiments using both real-world traces and synthetic traces to evaluate the proposed L4L approach. The results demonstrate the superiority of L4L as compared with the state-of-the-art solutions. [ABSTRACT FROM AUTHOR]

Steam flooding is an effective way to develop extra-heavy oil reservoir. However, when the steam flooding process enters into the exhaustion stage, considering lower heat utilization, higher water and poor development performance, how to enhance heavy oil recovery in the post steam flooding process is a large challenge. This paper provides various experiments studying on non-condensable gas-assisted steam flooding. Specifically, the effects of CO2 and N2 injection on the physical properties of extra-heavy oil were analyzed by High Temperature High Pressure (HTHP) PVT experiments. The effects of CO2 and N2 on the interfacial tension were studied using Axisymmetric Drop Shape Analysis (ADSA) system and the heat transfer characteristic of core which was saturated with N2 or water was studied with the sand pack model. Finally, different non-condensable gas (NCG)-assisted steam flooding experiments were conducted using the sand pack model. The results showed that the non-condensable gas dissolving into extra-heavy oil could effectively reduce the oil viscosity and density, and then increase the oil flow capability. The higher pressure and lower temperature was, the smaller the interfacial tension (IFT) was. Compared with N2, CO2 could significantly reduce the extra-heavy oil viscosity and IFT. But N2 could increase the sweep efficiency and heat utilization of steam. Compared with steam flooding process, non-condensable gas-assisted steam flooding process could prolong the anhydrous production period and increase the anhydrous recovery. Meanwhile, the non-condensable gas-assisted steam flooding process could increase the oil relative permeability and reduce residual oil saturation. The earlier implementing non-condensable gas-assisted steam flooding, the better the development performance would be. [ABSTRACT FROM AUTHOR]