Selecting an effective training signal for tasks in natural language processing is difficult: collecting expert annotations is expensive, and crowd-sourced annotations may not be reliable. At the same time, recent work in machine learning has demonstrated that learning from soft-labels acquired from crowd annotations can be effective, especially when there is distribution shift in the test set. However, the best method for acquiring these soft labels is inconsistent across tasks. This paper proposes new methods for acquiring soft-labels from crowd-annotations by aggregating the distributions produced by existing methods. In particular, we propose to find a distribution over classes by learning from multiple-views of crowd annotations via temperature scaling and finding the Jensen-Shannon centroid of their distributions. We demonstrate that using these aggregation methods leads to best or near-best performance across four NLP tasks on out-of-domain test sets, mitigating fluctuations in performance when using the constituent methods on their own. Additionally, these methods result in best or near-best uncertainty estimation across tasks. We argue that aggregating different views of crowd-annotations as soft-labels is an effective way to ensure performance which is as good or better than the best individual view, which is useful given the inconsistency in performance of the individual methods.

事后观察合理性是一种玩一般游戏的方法，该游戏规定了针对一组偏差的单个代理的无重格学习动态，并进一步描述了具有介导的平衡的多个代理商之间的共同理性行为。为了在依次的决策设置中发展事后理性学习，我们将行为偏差形式化为一般偏差，尊重广泛形式游戏的结构。将时间选择的概念整合到反事实遗憾的最小化（CFR）中，我们介绍了广泛的遗憾最小化（EFR）算法，该算法对于任何给定的行为偏差都具有与集合的复杂性紧密相关的计算相关的行为偏差。我们识别行为偏差子集，部分序列偏差类型，这些类型还包含先前研究的类型并导致长度中等的游戏中有效的EFR实例。此外，我们对基准游戏中不同偏差类型实例化的EFR进行了彻底的经验分析，我们发现更强大的类型通常会引起更好的性能。

在最近在两人，零和游戏中取得成功的驱动下，人工智能在游戏中的工作越来越重视产生基于平衡策略的算法。但是，这种方法在培养通用游戏或两个以上玩家的能力的玩家中的效果较小，而不是在两人游戏中的零和零游戏中。一个有吸引力的替代方法是考虑自适应算法，以确保相对于修改行为可以实现的方面的强劲表现。这种方法还导致了游戏理论分析，但是在关节学习动力学而不是均衡的代理行为引起的相关性游戏中。我们在一般的顺序决策环境中发展并倡导这一对学习的事后理性理性框架。为此，我们在广泛的游戏中重新检查了介导的平衡和偏差类型，从而获得了更完整的理解和解决过去的误解。我们提出了一组示例，说明了文献中每种平衡的独特优势和劣势，并证明没有可牵引的概念可以包含所有其他概念。这一探究线在与反事实遗憾最小化（CFR）家族中算法相对应的偏差和平衡类的定义中达到顶点，将它们与文献中的所有其他人联系起来。更详细地研究CFR进一步导致相关游戏中合理性的新递归定义，该定义以自然适用于后代评估的方式扩展了顺序合理性。

As language models (LMs) scale, they develop many novel behaviors, good and bad, exacerbating the need to evaluate how they behave. Prior work creates evaluations with crowdwork (which is time-consuming and expensive) or existing data sources (which are not always available). Here, we automatically generate evaluations with LMs. We explore approaches with varying amounts of human effort, from instructing LMs to write yes/no questions to making complex Winogender schemas with multiple stages of LM-based generation and filtering. Crowdworkers rate the examples as highly relevant and agree with 90-100% of labels, sometimes more so than corresponding human-written datasets. We generate 154 datasets and discover new cases of inverse scaling where LMs get worse with size. Larger LMs repeat back a dialog user's preferred answer ("sycophancy") and express greater desire to pursue concerning goals like resource acquisition and goal preservation. We also find some of the first examples of inverse scaling in RL from Human Feedback (RLHF), where more RLHF makes LMs worse. For example, RLHF makes LMs express stronger political views (on gun rights and immigration) and a greater desire to avoid shut down. Overall, LM-written evaluations are high-quality and let us quickly discover many novel LM behaviors.

Massive data corpora like WebText, Wikipedia, Conceptual Captions, WebImageText, and LAION have propelled recent dramatic progress in AI. Large neural models trained on such datasets produce impressive results and top many of today's benchmarks. A notable omission within this family of large-scale datasets is 3D data. Despite considerable interest and potential applications in 3D vision, datasets of high-fidelity 3D models continue to be mid-sized with limited diversity of object categories. Addressing this gap, we present Objaverse 1.0, a large dataset of objects with 800K+ (and growing) 3D models with descriptive captions, tags, and animations. Objaverse improves upon present day 3D repositories in terms of scale, number of categories, and in the visual diversity of instances within a category. We demonstrate the large potential of Objaverse via four diverse applications: training generative 3D models, improving tail category segmentation on the LVIS benchmark, training open-vocabulary object-navigation models for Embodied AI, and creating a new benchmark for robustness analysis of vision models. Objaverse can open new directions for research and enable new applications across the field of AI.

As AI systems become more capable, we would like to enlist their help to supervise other AIs. We experiment with methods for training a harmless AI assistant through self-improvement, without any human labels identifying harmful outputs. The only human oversight is provided through a list of rules or principles, and so we refer to the method as 'Constitutional AI'. The process involves both a supervised learning and a reinforcement learning phase. In the supervised phase we sample from an initial model, then generate self-critiques and revisions, and then finetune the original model on revised responses. In the RL phase, we sample from the finetuned model, use a model to evaluate which of the two samples is better, and then train a preference model from this dataset of AI preferences. We then train with RL using the preference model as the reward signal, i.e. we use 'RL from AI Feedback' (RLAIF). As a result we are able to train a harmless but non-evasive AI assistant that engages with harmful queries by explaining its objections to them. Both the SL and RL methods can leverage chain-of-thought style reasoning to improve the human-judged performance and transparency of AI decision making. These methods make it possible to control AI behavior more precisely and with far fewer human labels.

Nonconvex optimization is central in solving many machine learning problems, in which block-wise structure is commonly encountered. In this work, we propose cyclic block coordinate methods for nonconvex optimization problems with non-asymptotic gradient norm guarantees. Our convergence analysis is based on a gradient Lipschitz condition with respect to a Mahalanobis norm, inspired by a recent progress on cyclic block coordinate methods. In deterministic settings, our convergence guarantee matches the guarantee of (full-gradient) gradient descent, but with the gradient Lipschitz constant being defined w.r.t.~the Mahalanobis norm. In stochastic settings, we use recursive variance reduction to decrease the per-iteration cost and match the arithmetic operation complexity of current optimal stochastic full-gradient methods, with a unified analysis for both finite-sum and infinite-sum cases. We further prove the faster, linear convergence of our methods when a Polyak-{\L}ojasiewicz (P{\L}) condition holds for the objective function. To the best of our knowledge, our work is the first to provide variance-reduced convergence guarantees for a cyclic block coordinate method. Our experimental results demonstrate the efficacy of the proposed variance-reduced cyclic scheme in training deep neural nets.

Photo-identification (photo-id) is one of the main non-invasive capture-recapture methods utilised by marine researchers for monitoring cetacean (dolphin, whale, and porpoise) populations. This method has historically been performed manually resulting in high workload and cost due to the vast number of images collected. Recently automated aids have been developed to help speed-up photo-id, although they are often disjoint in their processing and do not utilise all available identifying information. Work presented in this paper aims to create a fully automatic photo-id aid capable of providing most likely matches based on all available information without the need for data pre-processing such as cropping. This is achieved through a pipeline of computer vision models and post-processing techniques aimed at detecting cetaceans in unedited field imagery before passing them downstream for individual level catalogue matching. The system is capable of handling previously uncatalogued individuals and flagging these for investigation thanks to catalogue similarity comparison. We evaluate the system against multiple real-life photo-id catalogues, achieving mAP@IOU[0.5] = 0.91, 0.96 for the task of dorsal fin detection on catalogues from Tanzania and the UK respectively and 83.1, 97.5% top-10 accuracy for the task of individual classification on catalogues from the UK and USA.

We introduce a sketch-and-solve approach to speed up the Peng-Wei semidefinite relaxation of k-means clustering. When the data is appropriately separated we identify the k-means optimal clustering. Otherwise, our approach provides a high-confidence lower bound on the optimal k-means value. This lower bound is data-driven; it does not make any assumption on the data nor how it is generated. We provide code and an extensive set of numerical experiments where we use this approach to certify approximate optimality of clustering solutions obtained by k-means++.

The de facto standard of dynamic histogram binning for radiomic feature extraction leads to an elevated sensitivity to fluctuations in annotated regions. This may impact the majority of radiomic studies published recently and contribute to issues regarding poor reproducibility of radiomic-based machine learning that has led to significant efforts for data harmonization; however, we believe the issues highlighted here are comparatively neglected, but often remedied by choosing static binning. The field of radiomics has improved through the development of community standards and open-source libraries such as PyRadiomics. But differences in image acquisition, systematic differences between observers' annotations, and preprocessing steps still pose challenges. These can change the distribution of voxels altering extracted features and can be exacerbated with dynamic binning.