Session-based recommendation aims to predict the next item based on a user’s limited interactions within a short period.Existing approaches use mainly recurrent neural networks(RNNs)or graph neural networks(GNNs)to m...Session-based recommendation aims to predict the next item based on a user’s limited interactions within a short period.Existing approaches use mainly recurrent neural networks(RNNs)or graph neural networks(GNNs)to model the sequential patterns or the transition relationships between items.However,such models either ignore the over-smoothing issue of GNNs,or directly use cross-entropy loss with a softmax layer for model optimization,which easily results in the over-fitting problem.To tackle the above issues,we propose a self-supervised graph learning with target-adaptive masking(SGL-TM)method.Specifically,we first construct a global graph based on all involved sessions and subsequently capture the self-supervised signals from the global connections between items,which helps supervise the model in generating accurate representations of items in the ongoing session.After that,we calculate the main supervised loss by comparing the ground truth with the predicted scores of items adjusted by our designed target-adaptive masking module.Finally,we combine the main supervised component with the auxiliary self-supervision module to obtain the final loss for optimizing the model parameters.Extensive experimental results from two benchmark datasets,Gowalla and Diginetica,indicate that SGL-TM can outperform state-of-the-art baselines in terms of Recall@20 and MRR@20,especially in short sessions.展开更多
This work reports a comprehensive study of a novel polyol method that can successfully synthesize layered LiNi0.4Mn0.4Co0.2O2,spinel LiNi0.5Mn1.5O4,and olivine LiCoPO4 cathode materials.When properly designed,polyol m...This work reports a comprehensive study of a novel polyol method that can successfully synthesize layered LiNi0.4Mn0.4Co0.2O2,spinel LiNi0.5Mn1.5O4,and olivine LiCoPO4 cathode materials.When properly designed,polyol method offers many advantages such as low cost,ease of use,and proven scalability for industrial applications.Most importantly,the unique properties of polyol solvent allow for greater morphology control as shown by all the resulting materials exhibiting monodispersed nanoparticles morphology.This morphology contributes to improved lithium ion transport due to short diffusion lengths.Polyol-synthesized LiNi0.4Mn0.4Co0.2O2 delivers a reversible capacity of 101 and 82 mAh.g-1 using high current rate of 5C and 10C,respectively.It also displays surprisingly high surface structure stability after chargedischarge processes.Each step of the reaction was investigated to understand the underlying polyol synthesis mechanism.A combination of in situ and ex situ studies reveal the structural and chemical transformation of Ni-Co alloy nanocrystals overwrapped by a Mn-and Li-embedded organic matrix to a sedes of intermediate phases,and then eventually to the desired layered oxide phase with a homogeneous distribution of Ni,Co,and Mn.We envisage that this type of analysis will promote the development of optimized synthesis protocols by establishing links between experimental factors and important structural and chemical properties of the desired product.The insights can open a new direction of research to synthesize high-performance intercalation compounds by allowing unprecedented control of intermediate phases using experimental parameters.展开更多
文摘Session-based recommendation aims to predict the next item based on a user’s limited interactions within a short period.Existing approaches use mainly recurrent neural networks(RNNs)or graph neural networks(GNNs)to model the sequential patterns or the transition relationships between items.However,such models either ignore the over-smoothing issue of GNNs,or directly use cross-entropy loss with a softmax layer for model optimization,which easily results in the over-fitting problem.To tackle the above issues,we propose a self-supervised graph learning with target-adaptive masking(SGL-TM)method.Specifically,we first construct a global graph based on all involved sessions and subsequently capture the self-supervised signals from the global connections between items,which helps supervise the model in generating accurate representations of items in the ongoing session.After that,we calculate the main supervised loss by comparing the ground truth with the predicted scores of items adjusted by our designed target-adaptive masking module.Finally,we combine the main supervised component with the auxiliary self-supervision module to obtain the final loss for optimizing the model parameters.Extensive experimental results from two benchmark datasets,Gowalla and Diginetica,indicate that SGL-TM can outperform state-of-the-art baselines in terms of Recall@20 and MRR@20,especially in short sessions.
文摘This work reports a comprehensive study of a novel polyol method that can successfully synthesize layered LiNi0.4Mn0.4Co0.2O2,spinel LiNi0.5Mn1.5O4,and olivine LiCoPO4 cathode materials.When properly designed,polyol method offers many advantages such as low cost,ease of use,and proven scalability for industrial applications.Most importantly,the unique properties of polyol solvent allow for greater morphology control as shown by all the resulting materials exhibiting monodispersed nanoparticles morphology.This morphology contributes to improved lithium ion transport due to short diffusion lengths.Polyol-synthesized LiNi0.4Mn0.4Co0.2O2 delivers a reversible capacity of 101 and 82 mAh.g-1 using high current rate of 5C and 10C,respectively.It also displays surprisingly high surface structure stability after chargedischarge processes.Each step of the reaction was investigated to understand the underlying polyol synthesis mechanism.A combination of in situ and ex situ studies reveal the structural and chemical transformation of Ni-Co alloy nanocrystals overwrapped by a Mn-and Li-embedded organic matrix to a sedes of intermediate phases,and then eventually to the desired layered oxide phase with a homogeneous distribution of Ni,Co,and Mn.We envisage that this type of analysis will promote the development of optimized synthesis protocols by establishing links between experimental factors and important structural and chemical properties of the desired product.The insights can open a new direction of research to synthesize high-performance intercalation compounds by allowing unprecedented control of intermediate phases using experimental parameters.
