arXiv AI September 28, 2026 By Anugya A, Saket Mohanty, Abhilash Timmapur, Somya Rai

Insurance Reserve Intelligence Platform

The paper introduces an Insurance Reserve Intelligence Platform that blends a classical Thiele-equation solver with a Physics-Informed Neural Network (PINN) enhanced by Knowledge-Informed Neural Network (KINN) losses for term-life reserve modeling. It generates synthetic policies, calculates risk-adjusted premiums, and constructs reserve-ratio datasets, then trains a neural model using seven features to predict standardized reserve ratios, achieving high accuracy (R² = 0.9887) and a 119.53× speedup over the classical solver. The framework also supports sensitivity analysis, elasticity analysis, prototype optimization, and interest-rate scenario testing, while noting remaining challenges in monotonicity and out-of-distribution generalization.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

Jul 7

Deep Learning for Dynamic Programming with Recursive Utility

arXiv:2607. 04278v1 Announce Type: cross Abstract: We propose the first deep learning algorithm, the Certainty Equivalent Learning (CEL) algorithm, for solving high-dimensional discrete-time dynamic programming problems with recursive utility.

By Xianhua Peng, Wu Guo

benchmarks

arXiv Machine Learning

Aug 6

Adaptive Finite-Budget Training for CVaR Risk-Aware Q-Learning

arXiv:2608. 04305v1 Announce Type: new Abstract: Risk-aware Q-learning (RaQL) provides a model-free, two-timescale estimator for dynamic risk objectives, but its finite-budget behavior remains fragile: fixed inner-loop hyperparameters can produce unstable value estimates, persistent Bellman residuals, and inefficient sample reuse.

By Yifan Wu, Junjie Lei, Wenjie Huang

reinforcement-learning

arXiv Machine Learning

Jun 4

Loss-Conditional PINNs for Parametric PDE Families

arXiv:2606. 04420v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) approximate solutions of ODEs and PDEs by minimising a weighted combination of residual, boundary, initial, and data losses.

By Anna Lazareva, Alexander Tarakanov

arXiv AI

Sep 7

Artificial Intelligence in Equity and Crypto Markets: Progress, Profitability Evidence, and the Limits of Automated Investing

Artificial Intelligence now underpins investment workflows from data and prediction to execution and tool use, yet its technical prowess does not automatically translate into profitability. A comprehensive review of public research up to 31 August 2026 across equities, ETFs, crypto spot, perpetual futures, and on‑chain markets shows real progress in prediction, text processing, portfolio design, and workflow integration, but evidence for durable net performance remains thin. The study highlights that factors such as temporal contamination, survivorship bias, weak benchmarks, implementation costs, and venue mechanics can erode alpha, and no single AI architecture has proven to deliver persistent, cross‑regime, capacity‑aware net alpha. "whyItMatters":"The findings underscore that while AI advances are evident, investors must rigorously test and govern AI systems to avoid overestimating their profitability potential."

By Linsen Zhu, Mengqing Cai

llms agents reinforcement-learning benchmarks

arXiv Machine Learning

Sep 23

Financially Guided Deep Portfolio Optimization

arXiv:2605.28853v2 Announce Type: replace-cross Abstract: Portfolio optimization in real-world financial markets is notoriously difficult due to non-stationarity, noisy data, and high transaction cos...

By Rahul Fernandes, Travis Desell

rag

arXiv Machine Learning

Jun 15

Deep Learning and Elicitability for McKean-Vlasov FBSDEs With Common Noise

arXiv:2512. 14967v2 Announce Type: replace Abstract: We present a novel numerical method for solving McKean--Vlasov forward--backward stochastic differential equations (MV--FBSDEs) with common noise, combining Picard iterations, elicitability and deep learning.

By Felipe J. P. Antunes, Yuri F. Saporito, Sebastian Jaimungal

One email a morning, machine-written

Related stories

Deep Learning for Dynamic Programming with Recursive Utility

Adaptive Finite-Budget Training for CVaR Risk-Aware Q-Learning

Loss-Conditional PINNs for Parametric PDE Families

Artificial Intelligence in Equity and Crypto Markets: Progress, Profitability Evidence, and the Limits of Automated Investing

Financially Guided Deep Portfolio Optimization

Deep Learning and Elicitability for McKean-Vlasov FBSDEs With Common Noise