Results 31 to 40 of about 84,677 (290)

Fooling LIME and SHAP [PDF]

Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, 2020
As machine learning black boxes are increasingly being deployed in domains such as healthcare and criminal justice, there is growing emphasis on building tools and techniques for explaining these black boxes in an interpretable manner. Such explanations are being leveraged by domain experts to diagnose systematic errors and underlying biases of black ...
Slack, Dylan, Hilgard, Sophie, Jia, Emily, Singh, Sameer, Lakkaraju, Himabindu   +4 more
openaire   +2 more sources

Reward Shaping with Recurrent Neural Networks for Speeding up On-Line Policy Learning in Spoken Dialogue Systems

, 2015
Statistical spoken dialogue systems have the attractive property of being able to be optimised from data via interactions with real users. However in the reinforcement learning paradigm the dialogue manager (agent) often requires significant time to ...
Gasic, Milica, Mrksic, Nikola, Su, Pei-Hao, Vandyke, David, Wen, Tsung-Hsien, Young, Steve   +5 more
core   +1 more source

Beyond saliency: understanding convolutional neural networks from saliency prediction on layer-wise relevance propagation [PDF]

, 2018
Despite the tremendous achievements of deep convolutional neural networks (CNNs) in many computer vision tasks, understanding how they actually work remains a significant challenge.
Li, Heyi, Tian, Yunke, Mueller, Klaus, Chen, Xin   +3 more
core   +3 more sources

Comprehensive Profiling of N6‐methyladnosine (m6A) Readouts Reveals Novel m6A Readers That Regulate Human Embryonic Stem Cell Differentiation

Advanced Science, EarlyView.
This research deciphers the m6A transcriptome by profiling its sites and functional readout effects: from mRNA stability, translation to alternative splicing, across five different cell types. Machine learning model identifies novel m6A‐binding proteins DDX6 and FXR2 and novel m6A reader proteins FUBP3 and L1TD1.
Zhou Huang, Rucong Liu, Zibaguli Wubulikasimu, Wanqing Zhao, Jiaqi Huang, Jiaxuan Wang, Tianyuan Zhang, Rui Fan, Wei Kong, Qinghua Cui, Yang Li, Yuan Zhou   +11 more
wiley   +1 more source

Characterization of Spatial and Temporal Coupling of Digital Economy and Carbon Emission in Yangtze River Delta Urban Agglomerations and the Influence Factors by Integrating GWRF and SHAP

Redai dili
Against the strategic backdrop of "Digital-China" and the "Dual-Carbon" goals, the synergistic advancement of digital economy and carbon emission reduction is crucial for achieving high-quality, sustainable development.
Zhang Qianwei, Xi Guangliang
doaj   +1 more source

Machine learning-guided synthesis of advanced inorganic materials

, 2019
Synthesis of advanced inorganic materials with minimum number of trials is of paramount importance towards the acceleration of inorganic materials development.
Chouhan, Tushar, Golani, Prafful, Guan, Cuntai, Liu, Zheng, Lu, Yuhao, Tang, Bijun, Wang, Han, Xu, Manzhang, Xu, Quan, Zhou, Jiadong   +9 more
core   +1 more source

Single‐Cell Metabolic Imaging and Digital Scoring of Fat Tissue Remodeling by Label‐Free Metabolic Microscopy

Advanced Science, EarlyView.
Mid‐infrared optoacoustic microscopy (MiROM) acquires lipid‐ and protein‐ associated vibrational contrast in intact fat tissue without dyes, preserving native tissue architecture. Through lateral and axial segmentation, MiROM tracks intrinsic intracellular changes during postnatal remodeling. A quantitative spatial analysis tool (Q‐SAT) maps white‐ and
Myeongseop Kim, Constantin Berger, Alexander Wolf, Alina Peteranderl, Martin Klingenspor, Vasilis Ntziachristos, Yongguo Li, Miguel A. Pleitez   +7 more
wiley   +1 more source

Study on the correlation between rail station area vitality and built environment

Journal of Asian Architecture and Building Engineering
To enhance the vitality of metro station areas and support station area development and optimisation, this study investigates 284 metro stations in Guangzhou.
Wangyang Gui, Wenli Wu, Di Wu
doaj   +1 more source

Prediction of recurrence-free survival in patients with renal cell carcinoma and tumor thrombosis of the renal and inferior vena cava of levels I–II using an extended Cox model and machine learning methods

Наука и инновации в медицине
Aim – to compare the predictive accuracy of Cox regression and machine learning (ML) methods regarding recurrence-free survival in patients with locally advanced renal cell carcinoma after radical treatment.
Musabek K. Mirzabekov, Nikolai D. Tikhonskii, Mikhail I. Shkolnik, Oleg A. Bogomolov, Nina V. Trukhacheva   +4 more
doaj   +1 more source

Towards Trustable SHAP Scores

Proceedings of the AAAI Conference on Artificial Intelligence
SHAP scores represent the proposed use of the well-known Shapley values in eXplainable Artificial Intelligence (XAI). Recent work has shown that the exact computation of SHAP scores can produce unsatisfactory results. Concretely, for some ML models, SHAP scores will mislead with respect to relative feature influence.
Letoffe, Olivier, Huang, Xuanxiang, Marques-Silva, Joao   +2 more
openaire   +2 more sources