Search for: All records

We propose a new learning framework that captures the tiered structure of many real-world user-interaction applications, where the users can be divided into two groups based on their different tolerance on exploration risks and should be treated separately. In this setting, we simultaneously maintain two policies π^O and πE: π^O ("O" for "online") interacts with more risk-tolerant users from the first tier and minimizes regret by balancing exploration and exploitation as usual, while π^E ("E" for "exploit") exclusively focuses on exploitation for risk-averse users from the second tier utilizing the data collected so far. An important question is whether such a separation yields advantages over the standard online setting (i.e., π^E=π^O) for the risk-averse users. We individually consider the gap-independent vs. gap-dependent settings. For the former, we prove that the separation is indeed not beneficial from a minimax perspective. For the latter, we show that if choosing Pessimistic Value Iteration as the exploitation algorithm to produce π^E, we can achieve a constant regret for risk-averse users independent of the number of episodes K, which is in sharp contrast to the Ω(logK) regret for any online RL algorithms in the same setting, while the regret of π^O (almost) maintains its online regret optimality and does not need to compromise for the success of π^E. 

We consider off-policy evaluation (OPE) in Partially Observable Markov Decision Processes (POMDPs), where the evaluation policy depends only on observable variables and the behavior policy depends on unobservable latent variables. Existing works either assume no unmeasured confounders, or focus on settings where both the observation and the state spaces are tabular. In this work, we first propose novel identification methods for OPE in POMDPs with latent confounders, by introducing bridge functions that link the target policy’s value and the observed data distribution. We next propose minimax estimation methods for learning these bridge functions, and construct three estimators based on these estimated bridge functions, corresponding to a value function-based estimator, a marginalized importance sampling estimator, and a doubly-robust estimator. Our proposal permits general function approximation and is thus applicable to settings with continuous or large observation/state spaces. The nonasymptotic and asymptotic properties of the proposed estimators are investigated in detail. A Python implementation of our proposal is available at https://github.com/jiaweihhuang/ Confounded-POMDP-Exp. 

Plasmonic nanostructures possess broadly tunable optical properties with catalytically active surfaces. They offer new opportunities for achieving efficient solar-to-chemical energy conversion. Plasmonic metal–semiconductor heterostructures have attracted heightened interest due to their capability of generating energetic hot electrons that can be collected to facilitate chemical reactions. In this article, we present a detailed survey of recent examples of plasmonic metal–semiconductor heterostructures for hot-electron-driven photochemistry, including plasmonic metal–oxide, plasmonic metal–two-dimensional materials, and plasmonic metal–metal–organic frameworks. We conclude with a discussion on the remaining challenges in the field and an outlook regarding future opportunities for designing high-performance plasmonic metal–semiconductor heterostructures for photochemistry. 

Communicating and understanding climate induced environmental changes can be challenging, especially using traditional representations such as graphs, maps or photos. Immersive visualizations and experiences offer an intuitive, visceral approach to otherwise rather abstract concepts, but creating them scientifically is challenging. In this paper, we linked ecological modeling, procedural modeling, and virtual reality to provide an immersive experience of a future forest. We mapped current tree species composition in northern Wisconsin using the Forest Inventory and Analysis (FIA) data and then forecast forest change 50 years into the future under two climate scenarios using LANDIS-II, a spatially-explicit, mechanistic simulation model. We converted the model output (e.g., tree biomass) into parameters required for 3D visualizations with analytical modeling. Procedural rules allowed us to efficiently and reproducibly translate the parameters into a simulated forest. Data visualization, environment exploration, and information retrieval were realized using the Unreal Engine. A system evaluation with experts in ecology provided positive feedback and future topics for a comprehensive ecosystem visualization and analysis approach. Our approach to create visceral experiences of forests under climate change can facilitate communication among experts, policy-makers, and the general public. 

Plasmon‐mediated carrier transfer (PMCT) at metal–semiconductor heterojunctions has been extensively exploited to drive photochemical reactions, offering intriguing opportunities for solar photocatalysis. However, to date, most studies have been conducted using noble metals. Inexpensive materials capable of generating and transferring hot carriers for photocatalysis via PMCT have been rarely explored. Here, we demonstrate that the plasmon excitation of nickel induces the transfer of both hot electrons and holes from Ni to TiO₂in a rationally designed Ni–TiO₂heterostructure. Furthermore, it is discovered that the transferred hot electrons either occupy oxygen vacancies (V_O) or produce Ti³⁺on TiO₂, while the transferred hot holes are located on surface oxygens at TiO₂. Moreover, the transferred hot electrons are identified to play a primary role in driving the degradation of methylene blue (MB). Taken together, our results validate Ni as a promising low‐cost plasmonic material for prompting visible‐light photochemical reactions.

 

Plasmon‐mediated carrier transfer (PMCT) at metal–semiconductor heterojunctions has been extensively exploited to drive photochemical reactions, offering intriguing opportunities for solar photocatalysis. However, to date, most studies have been conducted using noble metals. Inexpensive materials capable of generating and transferring hot carriers for photocatalysis via PMCT have been rarely explored. Here, we demonstrate that the plasmon excitation of nickel induces the transfer of both hot electrons and holes from Ni to TiO₂in a rationally designed Ni–TiO₂heterostructure. Furthermore, it is discovered that the transferred hot electrons either occupy oxygen vacancies (V_O) or produce Ti³⁺on TiO₂, while the transferred hot holes are located on surface oxygens at TiO₂. Moreover, the transferred hot electrons are identified to play a primary role in driving the degradation of methylene blue (MB). Taken together, our results validate Ni as a promising low‐cost plasmonic material for prompting visible‐light photochemical reactions.