A Learnable Wavelet Transformer for Long-Short Equity Trading and Risk-Adjusted Return Optimization

Ali NematiMar 322 sec read42 views

Researchers introduced WaveLSFormer, a wavelet-based Transformer model for intraday trading that directly optimizes market-neutral long/short portfolios under risk constraints. This approach outperforms traditional models like MLP and LSTM in terms of cumulative returns and Sharpe ratio across various industries, offering content creators insights into advanced financial modeling techniques.

Read the full article at arXiv cs.LG (ML)

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.

Comments

Bi-directional digital twin prototype anchoring with multi-periodicity learning for few-shot fault diagnosis

Researchers propose a bi-directional digital twin prototype anchoring method with multi-periodicity learning to enhance few-shot fault diagnosis in in...Researchers propose a bi-directional digital twin prototype anchoring method with multi-periodicity learning to enhance few-shot fault diagnosis in industrial machinery, addressing the challenge of limited labeled data. This approach improves model a...

Ali Nemati

AI & Machine LearningMar 328 sec read

Inside India's 10-Billion-a-Month Payments Engine

The National Payments Corporation of India (NPCI) processes over 10 billion financial transactions monthly using advanced technologies like federated ...The National Payments Corporation of India (NPCI) processes over 10 billion financial transactions monthly using advanced technologies like federated learning, graph neural networks, specialized LLMs, computer vision for license plate recognition, an...

Ali Nemati

Education & EdTechFeb 2725 sec read

Why Not Ask Why: Neuroscientist Urges Educators to Reconsider Technology's Reach

Neuroscientist Jared Cooney Horvath's book "The Digital Delusion" questions the overuse of technology in education and its impact on cognitive develop...Neuroscientist Jared Cooney Horvath's book "The Digital Delusion" questions the overuse of technology in education and its impact on cognitive development, arguing that screen time interferes with play-based learning essential for young children. He ...

Ali Nemati

AI & Machine LearningFeb 2524 sec read

CREDIT: Certified Ownership Verification of Deep Neural Networks Against Model Extraction Attacks

Researchers introduced CREDIT, a method to verify ownership of deep neural network models against model extraction attacks, which are common in MLaaS ...Researchers introduced CREDIT, a method to verify ownership of deep neural network models against model extraction attacks, which are common in MLaaS services. By quantifying model similarity using mutual information and setting verification threshol...

Ali Nemati

AI & Machine LearningFeb 2427 sec read

Beverly Hills-based Subject, whose streaming-style, on-demand platform provides digital curriculum and learning insights for grades 6-12, raised $28M (Natalie Breymeyer/Axios)

Subject, an AI-driven educational platform based in Beverly Hills, raised $28 million in a venture round led by Vistara Growth to support its streamin...Subject, an AI-driven educational platform based in Beverly Hills, raised $28 million in a venture round led by Vistara Growth to support its streaming-style, on-demand digital curriculum and learning insights for middle and high school students. Thi...

Ali Nemati

A Learnable Wavelet Transformer for Long-Short Equity Trading and Risk-Adjusted Return Optimization

Related Articles

Bi-directional digital twin prototype anchoring with multi-periodicity learning for few-shot fault diagnosis

Inside India's 10-Billion-a-Month Payments Engine

Why Not Ask Why: Neuroscientist Urges Educators to Reconsider Technology's Reach

CREDIT: Certified Ownership Verification of Deep Neural Networks Against Model Extraction Attacks

Beverly Hills-based Subject, whose streaming-style, on-demand platform provides digital curriculum and learning insights for grades 6-12, raised $28M (Natalie Breymeyer/Axios)