尽管在整个科学和工程中都无处不在，但只有少数部分微分方程（PDE）具有分析或封闭形式的解决方案。这激发了有关PDE的数值模拟的大量经典工作，最近，对数据驱动技术的研究旋转了机器学习（ML）。最近的一项工作表明，与机器学习的经典数值技术的混合体可以对任何一种方法提供重大改进。在这项工作中，我们表明，在纳入基于物理学的先验时，数值方案的选择至关重要。我们以基于傅立叶的光谱方法为基础，这些光谱方法比其他数值方案要高得多，以模拟使用平滑且周期性解决方案的PDE。具体而言，我们为流体动力学的三个模型PDE开发了ML增强的光谱求解器，从而提高了标准光谱求解器在相同分辨率下的准确性。我们还展示了一些关键设计原则，用于将机器学习和用于解决PDE的数值方法结合使用。

合奏是一种直接，非常有效的方法，用于提高模型在分类任务上的准确性，校准和鲁棒性；然而，其成功基础的原因仍然是研究的积极领域。我们基于PFAU（2013）的偏见变化分解的扩展，以便对分类器合奏的行为产生关键的见解。为了引入偏见变化权衡的双重重新聚集，我们首先得出了典型的分类任务的非对称损失的总期望和差异的广义定律。比较条件和引导偏置/方差估计值，我们表明条件估计必定会导致不可还原误差。接下来，我们表明在双空间中结合会降低差异并使偏差不变，而标准结合可以任意影响偏见。从经验上讲，标准的结合减少偏见，使我们假设分类器的集合可能会出现很好的表现，部分原因是这种意外的减少。我们通过对最近的深度学习方法的经验分析来结束，这些方法是在超级范围上进行整体，这表明这些技术确实有利于降低偏见偏见的偏见偏见。这表明，与经典智慧相反，靶向偏见可能是分类器合奏的有希望的方向。

Large "instruction-tuned" language models (finetuned to respond to instructions) have demonstrated a remarkable ability to generalize zero-shot to new tasks. Nevertheless, they depend heavily on human-written instruction data that is limited in quantity, diversity, and creativity, therefore hindering the generality of the tuned model. We introduce Self-Instruct, a framework for improving the instruction-following capabilities of pretrained language models by bootstrapping off its own generations. Our pipeline generates instruction, input, and output samples from a language model, then prunes them before using them to finetune the original model. Applying our method to vanilla GPT3, we demonstrate a 33% absolute improvement over the original model on Super-NaturalInstructions, on par with the performance of InstructGPT_001, which is trained with private user data and human annotations. For further evaluation, we curate a set of expert-written instructions for novel tasks, and show through human evaluation that tuning GPT3 with Self-Instruct outperforms using existing public instruction datasets by a large margin, leaving only a 5% absolute gap behind InstructGPT_001. Self-Instruct provides an almost annotation-free method for aligning pre-trained language models with instructions, and we release our large synthetic dataset to facilitate future studies on instruction tuning.

We introduce INSTRUCTOR, a new method for computing text embeddings given task instructions: every text input is embedded together with instructions explaining the use case (e.g., task and domain descriptions). Unlike encoders from prior work that are more specialized, INSTRUCTOR is a single embedder that can generate text embeddings tailored to different downstream tasks and domains, without any further training. We first annotate instructions for 330 diverse tasks and train INSTRUCTOR on this multitask mixture with a contrastive loss. We evaluate INSTRUCTOR on 70 embedding evaluation tasks (66 of which are unseen during training), ranging from classification and information retrieval to semantic textual similarity and text generation evaluation. INSTRUCTOR, while having an order of magnitude fewer parameters than the previous best model, achieves state-of-the-art performance, with an average improvement of 3.4% compared to the previous best results on the 70 diverse datasets. Our analysis suggests that INSTRUCTOR is robust to changes in instructions, and that instruction finetuning mitigates the challenge of training a single model on diverse datasets.

To train deep learning models, which often outperform traditional approaches, large datasets of a specified medium, e.g., images, are used in numerous areas. However, for light field-specific machine learning tasks, there is a lack of such available datasets. Therefore, we create our own light field datasets, which have great potential for a variety of applications due to the abundance of information in light fields compared to singular images. Using the Unity and C# frameworks, we develop a novel approach for generating large, scalable, and reproducible light field datasets based on customizable hardware configurations to accelerate light field deep learning research.

This article presents a survey of literature in the area of Human-Robot Interaction (HRI), specifically on systems containing more than two agents (i.e., having multiple humans and/or multiple robots). We identify three core aspects of ``Multi-agent" HRI systems that are useful for understanding how these systems differ from dyadic systems and from one another. These are the Team structure, Interaction style among agents, and the system's Computational characteristics. Under these core aspects, we present five attributes of HRI systems, namely Team size, Team composition, Interaction model, Communication modalities, and Robot control. These attributes are used to characterize and distinguish one system from another. We populate resulting categories with examples from recent literature along with a brief discussion of their applications and analyze how these attributes differ from the case of dyadic human-robot systems. We summarize key observations from the current literature, and identify challenges and promising areas for future research in this domain. In order to realize the vision of robots being part of the society and interacting seamlessly with humans, there is a need to expand research on multi-human -- multi-robot systems. Not only do these systems require coordination among several agents, they also involve multi-agent and indirect interactions which are absent from dyadic HRI systems. Adding multiple agents in HRI systems requires advanced interaction schemes, behavior understanding and control methods to allow natural interactions among humans and robots. In addition, research on human behavioral understanding in mixed human-robot teams also requires more attention. This will help formulate and implement effective robot control policies in HRI systems with large numbers of heterogeneous robots and humans; a team composition reflecting many real-world scenarios.

Language models can be prompted to perform a wide variety of zero- and few-shot learning problems. However, performance varies significantly with the choice of prompt, and we do not yet understand why this happens or how to pick the best prompts. In this work, we analyze the factors that contribute to this variance and establish a new empirical hypothesis: the performance of a prompt is coupled with the extent to which the model is familiar with the language it contains. Over a wide range of tasks, we show that the lower the perplexity of the prompt is, the better the prompt is able to perform the task. As a result, we devise a method for creating prompts: (1) automatically extend a small seed set of manually written prompts by paraphrasing using GPT3 and backtranslation and (2) choose the lowest perplexity prompts to get significant gains in performance.

Acquiring a better understanding of drought impacts becomes increasingly vital under a warming climate. Traditional drought indices describe mainly biophysical variables and not impacts on social, economic, and environmental systems. We utilized natural language processing and bidirectional encoder representation from Transformers (BERT) based transfer learning to fine-tune the model on the data from the news-based Drought Impact Report (DIR) and then apply it to recognize seven types of drought impacts based on the filtered Twitter data from the United States. Our model achieved a satisfying macro-F1 score of 0.89 on the DIR test set. The model was then applied to California tweets and validated with keyword-based labels. The macro-F1 score was 0.58. However, due to the limitation of keywords, we also spot-checked tweets with controversial labels. 83.5% of BERT labels were correct compared to the keyword labels. Overall, the fine-tuned BERT-based recognizer provided proper predictions and valuable information on drought impacts. The interpretation and analysis of the model were consistent with experiential domain expertise.

Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents to generate answers. Retrievers and readers are usually modeled separately, which necessitates a cumbersome implementation and is hard to train and adapt in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs Retrieval as Attention (ReAtt), and end-to-end training solely based on supervision from the end QA task. We demonstrate for the first time that a single model trained end-to-end can achieve both competitive retrieval and QA performance, matching or slightly outperforming state-of-the-art separately trained retrievers and readers. Moreover, end-to-end adaptation significantly boosts its performance on out-of-domain datasets in both supervised and unsupervised settings, making our model a simple and adaptable solution for knowledge-intensive tasks. Code and models are available at https://github.com/jzbjyb/ReAtt.

Language models trained on massive prompted multitask datasets like T0 (Sanh et al., 2021) or FLAN (Wei et al., 2021a) can generalize to tasks unseen during training. We show that training on a carefully chosen subset of instances can outperform training on all available data on a variety of datasets. We assume access to a small number (250--1000) of unlabeled target task instances, select their nearest neighbors from a pool of multitask data, and use the retrieved data to train target task-specific models. Our method is more data-efficient than training a single multitask model, while still outperforming it by large margins. We evaluate across a diverse set of tasks not in the multitask pool we retrieve from, including those used to evaluate T0 and additional complex tasks including legal and scientific document QA. We retrieve small subsets of P3 (the collection of prompted datasets from which T0's training data was sampled) and finetune T5 models that outperform the 3-billion parameter variant of T0 (T0-3B) by 3--30% on 12 out of 14 evaluation datasets while using at most 2% of the data used to train T0-3B. These models also provide a better initialization than T0-3B for few-shot finetuning on target-task data, as shown by a 2--23% relative improvement over few-shot finetuned T0-3B models on 8 datasets. Our code is available at https://github.com/allenai/data-efficient-finetuning.