Mladen Kolar from the University of Chicago Booth School of Business discussed stochastic optimization with equality constraints at MBZUAI. He presented a stochastic algorithm based on sequential quadratic programming (SQP) using a differentiable exact augmented Lagrangian. The algorithm adapts random stepsizes using a stochastic line search procedure, establishing global "almost sure" convergence. Why it matters: The presentation highlights MBZUAI's role in hosting discussions on advanced optimization techniques, fostering research and knowledge exchange in the field of machine learning.
A Marie Curie Fellow from Inria and UIUC presented research on stochastic gradient descent (SGD) through the lens of Markov processes, exploring the relationships between heavy-tailed distributions, generalization error, and algorithmic stability. The research challenges existing theories about the monotonic relationship between heavy tails and generalization error. It introduces a unified approach for proving Wasserstein stability bounds in stochastic optimization, applicable to convex and non-convex losses. Why it matters: The work provides novel insights into the theoretical underpinnings of stochastic optimization, relevant to researchers at MBZUAI and other institutions in the region working on machine learning algorithms.
KAUST hosted the Advances in Uncertainty Quantification Methods, Algorithms and Applications conference (UQAW2016) in January 2016. The event featured 75 presentations and 20 invited speakers from various countries. Professor Raul Tempone presented research on computational approaches to fouling accumulation and wear degradation using stochastic differential equations. Why it matters: This work provides a new computational approach based on stochastic differential equations to predict fouling patterns of heat exchangers which can optimize maintenance operations and reduce engine shut-down periods.
Alexander Gasnikov from the Moscow Institute of Physics and Technology presented a talk on open problems in convex optimization. The talk covered stochastic averaging vs stochastic average approximation, saddle-point problems and accelerated methods, homogeneous federated learning, and decentralized optimization. Gasnikov's research focuses on optimization algorithms and he has published in NeurIPS, ICML, EJOR, OMS, and JOTA. Why it matters: While the talk itself isn't directly related to GCC AI, understanding convex optimization is crucial for advancing machine learning algorithms used in the region.
Patrick van der Smagt, Director of AI Research at Volkswagen Group, discussed the use of generative machine learning models for predicting and controlling complex stochastic systems in robotics. The talk highlighted examples in robotics and beyond and addressed the challenges of achieving quality and trust in AI systems. He also mentioned his involvement in a European industry initiative on trust in AI and his membership in the AI Council of the State of Bavaria. Why it matters: Understanding control in robotics, along with trust in AI, are key issues for further development of autonomous systems, especially in industrial applications within the GCC region.
KAUST's Stochastic Numerics Research Group is developing methods for pricing European options. Their approach, detailed in an upcoming Journal of Computational Finance article, focuses on systematically tuning parameters to achieve accuracy while minimizing computational effort. The goal is to enable automated computation of fair prices for options contracts, similar to how insurance companies determine premiums. Why it matters: This research advances computational finance in the region, potentially improving risk management and investment strategies.
KAUST Professor Peter Richtárik received a Distinguished Speaker Award at the Sixth International Conference on Continuous Optimization (ICCOPT 2019) in Berlin. Richtárik's lecture series, totaling six hours, focused on stochastic gradient descent (SGD) methods, drawing from recent research by his KAUST group. He highlighted key principles and new variants of SGD, the key method for training modern machine learning models. Why it matters: This award recognizes KAUST's contribution to fundamental machine learning optimization, which is critical for advancing AI in the region.
This paper addresses exploration in reinforcement learning (RL) in unknown environments with sparse rewards, focusing on maximum entropy exploration. It introduces a game-theoretic algorithm for visitation entropy maximization with improved sample complexity of O(H^3S^2A/ε^2). For trajectory entropy, the paper presents an algorithm with O(poly(S, A, H)/ε) complexity, showing the statistical advantage of regularized MDPs for exploration. Why it matters: The research offers new techniques to reduce the sample complexity of RL, potentially enhancing the efficiency of AI agents in complex environments.