Genetic algorithm based deep learning model adaptation for improvising the motor imagery classification by R, Vishnupriya, Robinson, Neethu, M, Ramasubba Reddy

“…Transfer learning methods such as adaptation or fine-tuning can be used on the pre-trained model to improve the performance. …”
Get full text

Punching shear strength analysis of corroded reinforcement concrete slab by Fu, Yibin

“…Sea levels are rising due to climate change, hence this parameter is of increasing importance for structures under long term corrosive environment. There will be a significant decline in the shear strength resistance in the long run. …”
Get full text

Fluidity by Yan, Ran

“…Fluidity is a sustainable gender-fluid clothing collection, produced by using recycled hotel linen and natural dyeing methods. This project aims to challenge social norms and blur gender boundaries, providing modern consumers a combination of inclusivity and comfort through gender-fluid pieces.…”
Get full text

Plasma enhanced chemical vapor deposition growth of vanadium dioxide by LEE, Boon Peng

“…To achieve high quality VO2 thin film on a soda lime glass substrate, understanding nucleation mechanism to fine tune the process condition is essential. Are discussed in this report. …”
Get full text

SimSC: a simple framework for semantic correspondence with temperature learning by Li, X, Han, K, Wan, X, Prisacariu, VA

“…We discover that when fine-tuning ImageNet pre-trained backbone on the semantic matching task, L2 normalization of the feature map, a standard procedure in feature matching, produces an overly smooth matching distribution and significantly hinders the fine-tuning process. …”

Field instrumentation for performance assessment of Geobarrier System by Rahardjo, Harianto, Qian, Zhai, Satyanaga, Alfrendo, Leong, Eng Choon, Wang, Chien Looi, Wong, Johnny Liang Heng

“…The saturated hydraulic conductivity of a coarse-grained material is normally higher than that of a fine-grained material. However, the unsaturated hydraulic conductivity of the coarse-grained material decreases much faster than that of the fine-grained material with increasing matric suction and consequently the coarse-grained material can be less permeable than the fine-grained material in unsaturated condition. …”
Get full text
Get full text

Using AI for music source separation by Lee, Jasline Jie Yu

“…The objective is to analyse the impacts of different components present in both Spectrogram and Waveform based systems through fine-tuning, data handling and ablation testing. …”
Get full text

Cycle-based trading & portfolio management system by Zhan, Xiaoying

“…Thus, the target horizon is mid to long term. By selecting stocks at their troughs and investing capitals during the rising phases, capitals could be utilized more efficiently to preserve values and generate returns. …”
Get full text

Dynamic traction of lattice-confined platinum atoms into mesoporous carbon matrix for hydrogen evolution reaction by Lou, David Xiong Wen, Zhang, Huabin, An, Pengfei, Zhou, Wei, Guan, Bu Yuan, Zhang, Peng, Dong, Juncai

“…Results of extended x-ray absorption fine structure investigation and density functional theory calculation suggest that the active sites are associated with the lattice-confined Pt centers and the activated carbon (C)/nitrogen (N) atoms at the adjacency of the isolated Pt centers. …”
Get full text
Get full text

AI-powered COVID-19 forecasting: a comprehensive comparison of advanced deep learning methods by Tariq, Muhammad Usman, Ismail, Shuhaida

“…Bayesian optimization was also implemented to fine-tune these models. Results: The evaluation framework revealed that each model exhibited different levels of predictive accuracy and precision. …”
Get full text

AI-powered COVID-19 forecasting: a comprehensive comparison of advanced deep learning methods by Tariq, Muhammad Usman, Ismail, Shuhaida

“…Bayesian optimization was also implemented to fine-tune these models. Results: The evaluation framework revealed that each model exhibited different levels of predictive accuracy and precision. …”
Get full text

Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-level Backdoor Attacks by Zhang, Zhengyan, Xiao, Guangxuan, Li, Yongwei, Lv, Tian, Qi, Fanchao, Liu, Zhiyuan, Wang, Yasheng, Jiang, Xin, Sun, Maosong

“…The pre-training-then-fine-tuning paradigm has been widely used in deep learning. …”
Get full text

Build a Large Language Model (From Scratch) / by Raschka, Sebastian, author 656094

Get full text
Get full text

Technological transmission of knowledge in Neolithic northwestern China: mineralogical and chemical analyses of Yangshao and Majiayao painted ware by Spataro, M, Hein, A

“…The analyses were carried out to detect differences or similarities in the chaînes opératoires of ceramic production at the three sites, in particular, to examine the clay types selected to make the three main categories of wares for which Yangshao and Majiayao are known, namely painted fine wares, burnished fine wares, and unpainted coarse wares. …”

WiFi and vision multimodal self-supervised learning for human activity recognition by Tang, Shijie

“…In this context, we propose a novel Multimodal Pretrain fine-tune (MPF) network architecture for device-free human activity recognition using commercial WiFi-enabled IoT devices and cameras. …”
Get full text

Utilization of oil contaminated sand for green asphalt concrete production by Ng, Ye Han

“…The objective of this research is to study the feasibility of Oil Contaminated Sand (OCS) usage as a replacement of certain percentage of fine aggregates in Hot Mix Asphalt (HMA) for road construction. …”
Get full text

Properties of copper slag concrete by Zhang, Weide

“…This report investigated the mechanical properties of high strength concrete incorporating copper slag as fine aggregates. The rising need for alternative fine aggregates substitution of sand is essential due to its high materials cost and the emphasis on sustainable concrete. …”
Get full text

Structure and phase engineering of 2D transition metal chalcogenides by Tang, Bijun

“…As an indispensable member of the two-dimensional (2D) family as well as a perfect complement to graphene, 2D transition metal chalcogenides (TMCs) have long been the center of research attention, attributed to their large spectrum of fascinating properties along with the various crystal structures. …”
Get full text

Neural abstractive summarization: improvements at the sequence-level by Ravaut, Mathieu

“…With enough data, pre-trained Transformer-based sequence-to-sequence models can be fine-tuned and achieve strong performance on abstractive summarization benchmarks. …”
Get full text

Leveraging linguistic knowledge to enhance low-resource NLP applications by Zhu, Zixiao

“…Specifically, 1. In the promptless fine-tuning paradigm, we first focus on fine-tuning word embeddings for task-related words, thereby enriching the conceptual knowledge available to compositional neural networks during feature learning in emotion recognition. …”
Get full text