Search

Large language models (LLMs) have demonstrated remarkable zero-shot generalization abilities: state-of-the-art chatbots can provide plausible answers to many common questions that arise in daily life. However, so far, LLMs cannot reliably solve long-horizon planning problems. By contrast, classical planners, once a problem is given in a formatted way, can use efficient search algorithms to quickly identify correct, or even optimal, plans. In an effort to get the best of both worlds, this paper introduces LLM+P, the first framework that incorporates the strengths of classical planners into LLMs. LLM+P takes in a natural language description of a planning problem, then returns a correct (or optimal) plan for solving that problem in natural language. LLM+P does so by first converting the language description into a file written in the planning domain definition language (PDDL), then leveraging classical planners to quickly find a solution, and then translating the found solution back into natural language. Along with LLM+P, we define a diverse set of different benchmark problems taken from common planning scenarios. Via a comprehensive set of experiments on these benchmark problems, we find that LLM+P is able to provide optimal solutions for most problems, while LLMs fail to provide even feasible plans for most problems.\footnote{The code and results are publicly available at https://github.com/Cranial-XIX/llm-pddl.git.

In continuing tasks, average-reward reinforcement learning may be a more appropriate problem formulation than the more common discounted reward formulation. As usual, learning an optimal policy in this setting typically requires a large amount of training experiences. Reward shaping is a common approach for incorporating domain knowledge into reinforcement learning in order to speed up convergence to an optimal policy. However, to the best of our knowledge, the theoretical properties of reward shaping have thus far only been established in the discounted setting. This paper presents the first reward shaping framework for average-reward learning and proves that, under standard assumptions, the optimal policy under the original reward function can be recovered. In order to avoid the need for manual construction of the shaping function, we introduce a method for utilizing domain knowledge expressed as a temporal logic formula. The formula is automatically translated to a shaping function that provides additional reward throughout the learning process. We evaluate the proposed method on three continuing tasks. In all cases, shaping speeds up the average-reward learning rate without any reduction in the performance of the learned policy compared to relevant baselines.

Ultramicropores (size < 0.7 nm) are critically demanded to provide an efficient path for the penetration and transportation of electrolytes to achieve high-performance supercapacitors. Here, a self-sacrificial template approach is adopted, which introduces C8 alkyl chains with a kinetic diameter of 0.8-1 nm to occupy the cavity of a porous aromatic framework (PAF). During the heating process, the alkyl chains decompose from the dense architecture as the temperature increased from 500 to 600 °C, forming ∼1 nm micropores. The newly-obtained cavities provide sites for thermal-driven skeleton engineering (700-900 °C) to obtain ultramicropores. Based on the well-defined pore structure, the carbonized PAF solid revealed outstanding electrochemical performances, including high rate and long-term stability in a 6 M KOH electrolyte. Notably, the specific capacitance (294 F g-1) derived from the self-sacrificial template method exceeds the capability of all the other methods for the construction of ultramicropores including self-template strategy, carbonization of nanoparticles, and template-assisted strategy. The synthesis of ultramicroporous carbons via the self-sacrificial template route opens up a promising gate to adjust the porous structure for high-performance applications in supercapacitors.

Additional file 1: Figure S1. Routine mycoplasma test of engineered aMHC-mCherry H9 hPSCs. Representative PCR gel image of tested negative control (NC), inhibition control (IC), positive control (PC) and sample (S) were shown.

Additional file 2: Figure S2. Immunostaining against MLC2v (A) or MLC2a (B) with day 30 aMHC-mCherry CMs. Scale bar: 100 μm.

Additional file 3: Figure S3. Characterization of dual aMHC-mCherry and ASAP2f reporter H9 hPSCs. (A) Representative brightfield (BF) and ASAP2f images as well as flow cytometry analysis of live hPSCs were shown. (B) Representative immunostaining and flow cytometry analysis of NANOG, OCT4, and SSEA4 were shown. Scale bars, 100 μm.

Activation of p53-dependent apoptosis is critical for tumor suppression but aberrant activation of p53 also leads to developmental defects and heart failure. Here, we found that Rbm24 RNA-binding protein, a target of p53, regulates p53 mRNA translation. Mechanistically, we found that through binding to p53 mRNA and interaction with translation initiation factor eIF4E, Rbm24 prevents eIF4E from binding to p53 mRNA and inhibits the assembly of translation initiation complex. Importantly, we showed that mice deficient in Rbm24 die in utero due to the endocardial cushion defect in the heart at least in part due to aberrant activation of p53-dependent apoptosis. We also showed that the heart developmental defect in Rbm24-null mice can be partially rescued by p53 deficiency through decreased apoptosis in the heart. Together, we postulate that the p53-Rbm24 loop is critical for the heart development and may be explored for mitigating congenital heart diseases and heart failure.

Efforts are underway at UT Austin to build autonomous robot systems that address the challenges of long-term deployments in office environments and of the more prescribed domestic service tasks of the RoboCup@Home competition. We discuss the contrasts and synergies of these efforts, highlighting how our work to build a RoboCup@Home Domestic Standard Platform League entry led us to identify an integrated software architecture that could support both projects. Further, naturalistic deployments of our office robot platform as part of the Building-Wide Intelligence project have led us to identify and research new problems in a traditional laboratory setting., Comment: Presented at AI-HRI AAAI-FSS, 2018 (arXiv:1809.06606)

Task-motion planning (TMP) addresses the problem of efficiently generating executable and low-cost task plans in a discrete space such that the (initially unknown) action costs are determined by motion plans in a corresponding continuous space. However, a task-motion plan can be sensitive to unexpected domain uncertainty and changes, leading to suboptimal behaviors or execution failures. In this paper, we propose a novel framework, TMP-RL, which is an integration of TMP and reinforcement learning (RL) from the execution experience, to solve the problem of robust task-motion planning in dynamic and uncertain domains. TMP-RL features two nested planning-learning loops. In the inner TMP loop, the robot generates a low-cost, feasible task-motion plan by iteratively planning in the discrete space and updating relevant action costs evaluated by the motion planner in continuous space. In the outer loop, the plan is executed, and the robot learns from the execution experience via model-free RL, to further improve its task-motion plans. RL in the outer loop is more accurate to the current domain but also more expensive, and using less costly task and motion planning leads to a jump-start for learning in the real world. Our approach is evaluated on a mobile service robot conducting navigation tasks in an office area. Results show that TMP-RL approach significantly improves adaptability and robustness (in comparison to TMP methods) and leads to rapid convergence (in comparison to task planning (TP)-RL methods). We also show that TMP-RL can reuse learned values to smoothly adapt to new scenarios during long-term deployments.

When developing general purpose robots, the overarching software architecture can greatly affect the ease of accomplishing various tasks. Initial efforts to create unified robot systems in the 1990s led to hybrid architectures, emphasizing a hierarchy in which deliberative plans direct the use of reactive skills. However, since that time there has been significant progress in the low-level skills available to robots, including manipulation and perception, making it newly feasible to accomplish many more tasks in real-world domains. There is thus renewed optimism that robots will be able to perform a wide array of tasks while maintaining responsiveness to human operators. However, the top layer in traditional hybrid architectures, designed to achieve long-term goals, can make it difficult to react quickly to human interactions during goal-driven execution. To mitigate this difficulty, we propose a novel architecture that supports such transitions by adding a top-level reactive module which has flexible access to both reactive skills and a deliberative control module. To validate this architecture, we present a case study of its application on a domestic service robot platform., Comment: Presented at LTA AAAI-FSS, 2018

This paper addresses a novel architecture for person-following robots using active search. The proposed system can be applied in real-time to general mobile robots for learning features of a human, detecting and tracking, and finally navigating towards that person. To succeed at person-following, perception, planning, and robot behavior need to be integrated properly. Toward this end, an active target searching capability, including prediction and navigation toward vantage locations for finding human targets, is proposed. The proposed capability aims at improving the robustness and efficiency for tracking and following people under dynamic conditions such as crowded environments. A multi-modal sensor information approach including fusing an RGB-D sensor and a laser scanner, is pursued to robustly track and identify human targets. Bayesian filtering for keeping track of human and a regression algorithm to predict the trajectory of people are investigated. In order to make the robot autonomous, the proposed framework relies on a behavior-tree structure. Using Toyota Human Support Robot (HSR), real-time experiments demonstrate that the proposed architecture can generate fast, efficient person-following behaviors., Comment: 7 pages, 6 figures

控释肥作为一种能够提高肥料利用率、保障作物产量和节约劳动力的新型肥料已经在作物生产中得到广泛应用，而控释肥对土壤N2O排放影响结果的差异使其成为当前科学评估控释肥施用环境效应的焦点问题之一。因此，旨在探讨不同种类控释肥及氮素水平施用对华北平原冬小麦/夏玉米轮作系统土壤N2O排放的影响，为科学评价控释肥施用的环境效应及其推广应用提供科学依据。本研究监测采用静态暗箱-气相色谱法对不同控释肥施用下土壤N2O排放、环境因素以及产量进行了周年监测，探讨了不同处理（对照处理（CK）、控释肥处理1（CRF1）、优化控释肥处理1（80% CRF1）、优化控释肥处理2（80% CRF2）和控释肥处理3（CRF3+尿素））下土壤N2O排放特征及土壤温湿度对其的影响。结果表明：控释肥施用下冬小麦/夏玉米轮作系统中土壤N2O排放峰高值主要出现在基肥施用并伴随灌溉（或降雨）后，一般持续时间约为7-10 d，小麦返青期灌溉以及玉米后期降雨会引起微弱的N2O排放峰。不同处理土壤N2O排放通量变化范围为-235.61-2625.01 μg N2O m-2 h-1，平均排放通量为23.88-51.39 μg N2O m-2 h-1，与CRFI相比，80% CRF1和80% CRF2处理能够减小施肥期的N2O排放峰值，但不改变轮作周期土壤N2O排放季节变化规律。CK处理和CRF3+尿素处理土壤N2O排放通量与5 cm深度土壤温度之间表现出显著的正相关性（r2=0.38，P < 0.01；r2=0.30，P < 0.05）；CRF1处理和80% CRF1处理在冬小麦生长季及整个轮作周期内与土壤孔隙含水率（WFPS）表现为显著的正相关关系（冬小麦生长季分别为r2=0.50，P < 0.01；r2=0.39，P < 0.05；整个轮作周期分别为r2=0.39，P < 0.05；r2=0.43，P < 0.05）。80% CRF2处理N2O年排放总量最高，为（2.89±0.24）kg N/hm2。相同控释肥种类条件下，80% CRF1处理比CRF1处理减少了14.23%，但并未达到显著水平；相同施氮量水平下，CRF1处理与（CRF3+尿素）处理之间N2O年排放总量差异不显著，而80% CRF1处理比80% CRF2处理N2O年排放总量减少16.16%，并达到显著水平（P < 0.05）。本研究不同处理之间N2O直接排放系数在0.29%-0.42%之间，均明显低于IPCC 1.0%的默认值。各控释肥处理产量与当地农民常规施肥量条件下产量没有显著性差异。因此在华北地区冬小麦/夏玉米轮作系统中应用控释肥技术可以在保证产量的前提下有效减少土壤N2O排放，并且仍存在一定的减排空间。

Stable luminescent pi-radicals with doublet emission have aroused a growing interest for functional molecular materials. We have demonstrated a neutral pi-radical dye (4-N-carbazolyl-2,6-dichlorophenyl)bis(2,4,6-trichlorophenyl)-methyl (TTM-1Cz) with remarkable doublet emission, which could be used as triplet sensitizer to initiate the photophysical process of triplet-triplet annihilation photon upconversion (TTA-UC). Dexter-like excited doublet-triplet energy transfer (DTET) was confirmed by theoretical calculation. A mixed solution of TTM-1Cz and aromatic emitters could ambidextrous upconvert red light to cyan light or green light = 532 nm) to blue light. This finding of DTET phenomena provides a new perspective in designing new triplet sensitizer for TTA-UC., Comment: 22 pages, 12 figures