Science – Page 20

Spotting the Signals: Machine Learning’s Edge in Insider Trading Detection

12.11.2025 by qfx

The analysis ranked feature importance using a metric—mean decrease in impurity—calculated during the random forest training process, offering insight into which variables most influenced the model’s decisions as outlined in Neupane et al. (2024).

This study demonstrates how advanced machine learning techniques can effectively identify illicit trading activity, enhancing financial market surveillance.

The Algorithm Hunter: How AI is Starting to Write Its Own Code

12.11.2025 by qfx

AlphaResearch initiates a cycle of autonomous refinement, beginning with the training of reward models using established peer-reviewed research, and subsequently leveraging this foundation to generate and iteratively improve research proposals and evaluation programs—a system designed not to circumvent decay, but to manage it through continuous adaptation.

An autonomous AI agent, AlphaResearch, is pushing the boundaries of algorithm discovery by generating, testing, and refining code with minimal human intervention.

Mastering the Art of Deception: An AI Conquers Stratego

12.11.2025 by qfx

A new artificial intelligence, trained through self-play and strategic search, has achieved superhuman performance in the classic game of imperfect information.

Learning from Ghosts: Stabilizing Off-Policy Imitation

12.11.2025 by qfx

An off-policy actor-critic architecture facilitates continuous control, as demonstrated in the work of Liessner et al. (2018) on deep reinforcement learning for energy management.

A novel approach to imitation learning tackles the challenges of learning from previously collected data, dramatically improving sample efficiency.

From Paper to Data: Automating Invoice Processing with AI

11.11.2025 by qfx

This review explores how artificial intelligence is transforming invoice handling, moving beyond manual data entry.

Spotting the Fakes: A New Framework for AI-Generated Misinformation

11.11.2025 by qfx

As AI-created content floods social media, discerning truth from fabrication is becoming increasingly difficult.

The Fading Shield Against Deepfakes

11.11.2025 by qfx

New research highlights the performance decay of current deepfake detection models when trained on static datasets, emphasizing the need for continuous learning and improved feature robustness.

Splitting the Difference: Smarter Models from Complex AI

11.11.2025 by qfx

Through the decomposition of complex large language model features—achieved by varying three decoupling variables—specialized small models are distilled, demonstrating a pathway to simplify intricate systems into manageable, focused components.

A new approach decomposes the knowledge of large AI models into simpler components, boosting performance in financial applications.

Taming Volatility with Oscillating Networks

11.11.2025 by qfx

The system leverages the temporal dynamics of a Lee Oscillator, distilling its oscillatory behavior into a meta-activation function through Max-over-Time pooling—a process that suggests any attempt at static definition inevitably loses information inherent in the system’s evolving state.

A new deep learning approach leverages the power of chaotic oscillators to improve forecasting in unpredictable systems.

The AI Cooperation Problem

11.11.2025 by qfx

Can economic incentives prevent a catastrophic future with superintelligent artificial intelligence?