Prediction of Shanghai and Shenzhen 300 Index Based on BP Neural Network

Hong, Lu; Guo, Nan; Zhao, Bingbing; Wang, Yi-han; Xia, Fang

doi:10.1007/978-981-15-8462-6_4

Cited by 4 publications

(11 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) ODIN (Liang, Li, and Srikant 2017), which leverages temperature scaling and input perturbation to improve OOD detection. 3) Energy score (Liu et al 2020), which utilizes the information in logits for OOD detection, where logits represent the negative log of the denominator in the softmax function.…”

Section: Main Results and Discussionmentioning

confidence: 99%

“…Mahalanobis distancebased scores (Lee et al 2018) were used for OOD detection by modeling the class-conditional distributions of softmax neural classifiers using multivariate Gaussian distributions. Energy-OOD (Liu et al 2020) proposes using an energy score for OOD detection, which is theoretically more aligned with the probability density of the inputs and is less likely to result in overconfident predictions. ReAct (Sun, Guo, and Li 2021) rectifies feature vectors by thresholding their elements with a certain magnitude.…”

Section: Related Workmentioning

confidence: 99%

“…Numerous OOD detection methods (Hendrycks and Gimpel 2016;Liang, Li, and Srikant 2017;Lee et al 2018;Liu et al 2020;Huang, Geng, and Li 2021) have relied on employing off-the-shelf softmax cross-entropy (CE) loss to produce embeddings that are sufficient for classifying ID samples. However, this approach may not be optimally designed for OOD detection since it cannot guarantee high levels of intra-class compactness and inter-class separability (Liu et al 2016(Liu et al , 2017Wang et al 2018).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CASE: Exploiting Intra-class Compactness and Inter-class Separability of Feature Embeddings for Out-of-Distribution Detection

Feng,

Jin,

Wang

2024

AAAI

View full text Add to dashboard Cite

Detecting out-of-distribution (OOD) inputs is critical for reliable machine learning, but deep neural networks often make overconfident predictions, even for OOD inputs that deviate from the distribution of training data. Prior methods relied on the widely used softmax cross-entropy (CE) loss that is adequate for classifying in-distribution (ID) samples but not optimally designed for OOD detection. To address this issue, we propose CASE, a simple and effective OOD detection method by explicitly improving intra-class Compactness And inter-class Separability of feature Embeddings. To enhance the separation between ID and OOD samples, CASE uses a dual-loss framework, which includes a separability loss that maximizes the inter-class Euclidean distance to promote separability among different class centers, along with a compactness loss that minimizes the intra-class Euclidean distance to encourage samples to be close to their class centers. In particular, the class centers are defined as a free optimization parameter of the model and updated by gradient descent, which is simple and further enhances the OOD detection performance. Extensive experiments demonstrate the superiority of CASE, which reduces the average FPR95 by 37.11% and improves the average AUROC by 15.89% compared to the baseline method using a softmax confidence score on the more challenging CIFAR-100 model.

show abstract

Section: Main Results and Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CASE: Exploiting Intra-class Compactness and Inter-class Separability of Feature Embeddings for Out-of-Distribution Detection

Feng,

Jin,

Wang

2024

AAAI

View full text Add to dashboard Cite

show abstract

“…In this paper, we experimented with two different architectures to verify the impact of using different OOD detection methods on our framework. We conducted experiments using two methods, Softmax score (Hendrycks and Gimpel 2016) and Energy (Liu et al 2020), on the f-CLSWGAN baseline model. The experimental results in Table 4 showed that no matter which architecture was used, the performance was significantly improved.…”

Section: Ablation Studymentioning

confidence: 99%

“…OOD detection(Hendrycks and Gimpel 2016;Hendrycks, Mazeika, and Dietterich 2018;Lee et al 2017;Wang et al 2022) is the task of detecting when a sample is drawn from a distribution different from the training data. Some techniques(Huang, Geng, and Li 2021;Liang, Li, and Srikant 2017;Liu et al 2020;Hendrycks and Gimpel 2016;Lee et al 2018;Sun, Guo, and Li 2021) concentrate on developing…”

mentioning

confidence: 99%

Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition

Wang,

Hong,

Huangfu

et al. 2024

AAAI

View full text Add to dashboard Cite

In the realm of Zero-Shot Learning (ZSL), we address biases in Generalized Zero-Shot Learning (GZSL) models, which favor seen data. To counter this, we introduce an end-to-end generative GZSL framework called D3GZSL. This framework respects seen and synthesized unseen data as in-distribution and out-of-distribution data, respectively, for a more balanced model. D3GZSL comprises two core modules: in-distribution dual space distillation (ID2SD) and out-of-distribution batch distillation (O2DBD). ID2SD aligns teacher-student outcomes in embedding and label spaces, enhancing learning coherence. O2DBD introduces low-dimensional out-of-distribution representations per batch sample, capturing shared structures between seen and un seen categories. Our approach demonstrates its effectiveness across established GZSL benchmarks, seamlessly integrating into mainstream generative frameworks. Extensive experiments consistently showcase that D3GZSL elevates the performance of existing generative GZSL methods, under scoring its potential to refine zero-shot learning practices. The code is available at: https://github.com/PJBQ/D3GZSL.git

show abstract

Research on Intelligent Garbage Classification Algorithm Based on Deep Learning

Wang

2022

2022 International Conference on Artificial Intelligence, Information Processing and Cloud Computing (AIIPCC)

View full text Add to dashboard Cite

Prediction of Shanghai and Shenzhen 300 Index Based on BP Neural Network

Cited by 4 publications

References 10 publications

CASE: Exploiting Intra-class Compactness and Inter-class Separability of Feature Embeddings for Out-of-Distribution Detection

CASE: Exploiting Intra-class Compactness and Inter-class Separability of Feature Embeddings for Out-of-Distribution Detection

Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition

Research on Intelligent Garbage Classification Algorithm Based on Deep Learning

Contact Info

Product

Resources

About