Mansur Arief

The global energy transition requires increased production of critical minerals. However, selecting which target deposits to explore and subsequently to develop into active mines is a staged process: each exploration action generates new observations that reduce uncertainty and inform the next decision, whether to commit capital, delay, or abandon a target. Project viability is further constrained by socio-environmental factors that can also evolve as projects progress. Because decisions unfold incrementally, with each choice contingent on information from prior stages, this is inherently a sequential decision-making problem; we propose a Partially Observable Markov Decision Process (POMDP) framework to optimize mineral exploration and mine development strategies across multiple targets. Our model explicitly incorporates uncertainties in ore grade, tonnage, deposit depth, and commodity price, alongside a dynamic sustainability score that imposes escalating penalties when mining activities breach environmental and social license thresholds. By generating observations through exploration, the planner updates its belief state, weighing the value of new information against the options to proceed, delay, or abandon a target. We evaluate this belief-aware AI planner on synthetic multi-target scenarios, comparing its performance against a baseline AI planner that ignores early-stage sustainability metrics and a traditional multi-criteria decision-making (MCDM) agent that strictly maximizes Net Present Value. The results demonstrate that the proposed planner optimally balances profitability and risk, actively avoiding severe sustainability penalties even at the cost of abandoning high-grade or high-tonnage targets. Ultimately, this framework provides a strategic tool for aligning economic objectives with environmental stewardship and social acceptance.

Training intelligent agents to navigate highly interactive environments presents significant challenges. While guided meta reinforcement learning (RL) approach that first trains a guiding policy to train the ego agent has proven effective in improving generalizability across scenarios with various levels of interaction, the state-of-the-art method tends to be overly sensitive to extreme cases, impairing the agents' performance in the more common scenarios. This study introduces a novel training framework that integrates guided meta RL with importance sampling (IS) to optimize training distributions iteratively for navigating highly interactive driving scenarios, such as T-intersections or roundabouts. Unlike traditional methods that may underrepresent critical interactions or overemphasize extreme cases during training, our approach strategically adjusts the training distribution towards more challenging driving behaviors using IS proposal distributions and applies the importance ratio to de-bias the result. By estimating a naturalistic distribution from real-world datasets and employing a mixture model for iterative training refinements, the framework ensures a balanced focus across common and extreme driving scenarios. Experiments conducted with both synthetic and naturalistic datasets demonstrate both accelerated training and performance improvements under highly interactive driving tasks.

This study addresses the challenge of stroke diagnosis and treatment under uncertainty, a critical issue given the rapid progression and severe consequences of stroke conditions such as aneurysms, arteriovenous malformations (AVM), and occlusions. Current diagnostic methods, including Digital Subtraction Angiography (DSA), face limitations due to high costs and its invasive nature. To overcome these challenges, we propose a novel approach using a Partially Observable Markov Decision Process (POMDP) framework. Our model integrates advanced diagnostic tools and treatment approaches with a decision-making algorithm that accounts for the inherent uncertainties in stroke diagnosis. Our approach combines noisy observations from CT scans, Siriraj scores, and DSA reports to inform the subsequent treatment options. We utilize the online solver DESPOT, which employs tree-search methods and particle filters, to simulate potential future scenarios and guide our strategies. The results indicate that our POMDP framework balances diagnostic and treatment objectives, striking a tradeoff between the need for precise stroke identification via invasive procedures like DSA and the constraints of limited healthcare resources that necessitate more cost-effective strategies, such as in-hospital or at-home observation, by relying only relying on simulation rollouts and not imposing any prior knowledge. Our study offers a significant contribution by presenting a systematic framework that optimally integrates diagnostic and treatment processes for stroke and accounting for various uncertainties, thereby improving care and outcomes in stroke management.

Autonomous driving systems have witnessed a significant development during the past years thanks to the advance in machine learning-enabled sensing and decision-making algorithms. One critical challenge for their massive deployment in the real world is their safety evaluation. Most existing driving systems are still trained and evaluated on naturalistic scenarios collected from daily life or heuristically-generated adversarial ones. However, the large population of cars, in general, leads to an extremely low collision rate, indicating that the safety-critical scenarios are rare in the collected real-world data. Thus, methods to artificially generate scenarios become crucial to measure the risk and reduce the cost. In this survey, we focus on the algorithms of safety-critical scenario generation in autonomous driving. We first provide a comprehensive taxonomy of existing algorithms by dividing them into three categories: data-driven generation, adversarial generation, and knowledge-based generation. Then, we discuss useful tools for scenario generation, including simulation platforms and packages. Finally, we extend our discussion to five main challenges of current works -- fidelity, efficiency, diversity, transferability, controllability -- and research opportunities lighted up by these challenges.

Autonomous driving systems have witnessed significant development during the past years thanks to the advance in machine learning-enabled sensing and decision-making algorithms. One critical challenge for their massive deployment in the real world is their safety evaluation. Most existing driving systems are still trained and evaluated on naturalistic scenarios collected from daily life or heuristically-generated adversarial ones. However, the large population of cars, in general, leads to an extremely low collision rate, indicating that safety-critical scenarios are rare in the collected real-world data. Thus, methods to artificially generate scenarios become crucial to measure the risk and reduce the cost. In this survey, we focus on the algorithms of safety-critical scenario generation in autonomous driving. We first provide a comprehensive taxonomy of existing algorithms by dividing them into three categories: data-driven generation, adversarial generation, and knowledge-based generation. Then, we discuss useful tools for scenario generation, including simulation platforms and packages. Finally, we extend our discussion to five main challenges of current works – fidelity, efficiency, diversity, transferability, controllability – and research opportunities lighted up by these challenges.

The COVID-19 pandemic has recently exacerbated the fierce competition in the transportation businesses. The airline industry took one of the biggest hits as the closure of international borders forced aircraft operators to suspend their international routes, keeping aircraft on the ground without generating revenues while at the same time still requiring adequate maintenance. To maintain their operational sustainability, finding a good balance between cost reductions measure and safety standards fulfillment, including its maintenance procedure, becomes critical. This paper proposes an AI-assisted predictive maintenance scheme that synthesizes prognostics modeling and simulation-based optimization to help airlines decide their optimal engine maintenance approach. The proposed method enables airlines to utilize their diagnostics measurements and operational settings to design a more customized maintenance strategy that takes engine operations conditions into account. Our numerical experiments on the proposed approach resulted in significant cost savings without compromising the safety standards. The experiments also show that maintenance strategies tailored to the failure mode and operational settings (that our framework enables) yield 13% more cost savings than generic optimal maintenance strategies. The generality of our proposed framework allows the extension to other intelligent, safety-critical transportation systems.