Language agents have shown some ability to interact with an external environment, e.g., a virtual world such as ScienceWorld, to perform complex tasks, e.g., growing a plant, without the startup costs of reinforcement learning. However, despite their zero-shot capabilities, these agents to date do not continually improve over time beyond performance refinement on a specific task. Here we present CLIN, the first language-based agent to achieve this, so that it continually improves over multiple trials, including when both the environment and task are varied, and without requiring parameter updates. Our approach is to use a persistent, dynamic, textual memory centered on causal abstractions (rather than general "helpful hints") that is regularly updated after each trial so that the agent gradually learns useful knowledge for new trials. In the ScienceWorld benchmark, CLIN is able to continually improve on repeated trials on the same task and environment, outperforming state-of-the-art reflective language agents like Reflexion by 23 absolute points. CLIN can also transfer its learning to new environments (or new tasks), improving its zero-shot performance by 4 points (13 for new tasks) and can further improve performance there through continual memory updates, enhancing performance by an additional 17 points (7 for new tasks). This suggests a new architecture for agents built on frozen models that can still continually and rapidly improve over time.

Recently, SyncMap pioneered an approach to learn complex structures from sequences as well as adapt to any changes in underlying structures. This is achieved by using only nonlinear dynamical equations inspired by neuron group behaviors, i.e., without loss functions. Here we propose Symmetrical SyncMap that goes beyond the original work to show how to create dynamical equations and attractor-repeller points which are stable over the long run, even dealing with imbalanced continual general chunking problems (CGCPs). The main idea is to apply equal updates from negative and positive feedback loops by symmetrical activation. We then introduce the concept of memory window to allow for more positive updates. Our algorithm surpasses or ties other unsupervised state-of-the-art baselines in all 12 imbalanced CGCPs with various difficulties, including dynamically changing ones. To verify its performance in real-world scenarios, we conduct experiments on several well-studied structure learning problems. The proposed method surpasses substantially other methods in 3 out of 4 scenarios, suggesting that symmetrical activation plays a critical role in uncovering topological structures and even hierarchies encoded in temporal data.

A robot providing mealtime assistance must perform specialized maneuvers with various utensils in order to pick up and feed a range of food items. Beyond these dexterous low-level skills, an assistive robot must also plan these strategies in sequence over a long horizon to clear a plate and complete a meal. Previous methods in robot-assisted feeding introduce highly specialized primitives for food handling without a means to compose them together. Meanwhile, existing approaches to long-horizon manipulation lack the flexibility to embed highly specialized primitives into their frameworks. We propose Visual Action Planning OveR Sequences (VAPORS), a framework for long-horizon food acquisition. VAPORS learns a policy for high-level action selection by leveraging learned latent plate dynamics in simulation. To carry out sequential plans in the real world, VAPORS delegates action execution to visually parameterized primitives. We validate our approach on complex real-world acquisition trials involving noodle acquisition and bimanual scooping of jelly beans. Across 38 plates, VAPORS acquires much more efficiently than baselines, generalizes across realistic plate variations such as toppings and sauces, and qualitatively appeals to user feeding preferences in a survey conducted across 49 individuals. Code, datasets, videos, and supplementary materials can be found on our website: https://sites.google.com/view/vaporsbot.

Time series data are found in many areas of healthcare such as medical time series, electronic health records (EHR), measurements of vitals, and wearable devices. Causal discovery, which involves estimating causal relationships from observational data, holds the potential to play a significant role in extracting actionable insights about human health. In this study, we present a novel constraint-based causal discovery approach for autocorrelated and non-stationary time series data (CDANs). Our proposed method addresses several limitations of existing causal discovery methods for autocorrelated and non-stationary time series data, such as high dimensionality, the inability to identify lagged causal relationships, and overlooking changing modules. Our approach identifies lagged and instantaneous/contemporaneous causal relationships along with changing modules that vary over time. The method optimizes the conditioning sets in a constraint-based search by considering lagged parents instead of conditioning on the entire past that addresses high dimensionality. The changing modules are detected by considering both contemporaneous and lagged parents. The approach first detects the lagged adjacencies, then identifies the changing modules and contemporaneous adjacencies, and finally determines the causal direction. We extensively evaluated our proposed method on synthetic and real-world clinical datasets, and compared its performance with several baseline approaches. The experimental results demonstrate the effectiveness of the proposed method in detecting causal relationships and changing modules for autocorrelated and non-stationary time series data.

Density estimation, which estimates the distribution of data, is an important category of probabilistic machine learning. A family of density estimators is mixture models, such as Gaussian Mixture Model (GMM) by expectation maximization. Another family of density estimators is the generative models which generate data from input latent variables. One of the generative models is the Masked Autoregressive Flow (MAF) which makes use of normalizing flows and autoregressive networks. In this paper, we use the density estimators for classification, although they are often used for estimating the distribution of data. We model the likelihood of classes of data by density estimation, specifically using GMM and MAF. The proposed classifiers outperform simpler classifiers such as linear discriminant analysis which model the likelihood using only a single Gaussian distribution. This work opens the research door for proposing other probabilistic classifiers based on joint density estimation.

In this work, we consider a classification problem in unsupervised domain adaptation (UDA). UDA aims to transfer knowledge from a source domain with ample labeled data to enhance the performance in a target domain where labeled data is unavailable. In UDA, the source and target domains have different data generating distributions, so the core challenge is to transfer knowledge contained in the labeled dataset in the source domain to the target domain under the distribution shifts. Over the decades, significant improvements in the transferability from source to target domains have been made, resulting in areas like domain alignment (Ben-David et al., 2010; Ganin et al., 2016; Long et al., 2018; Zhang et al., 2019) and self-training (Cai et al., 2021; Chen et al., 2020; Liu et al., 2021). Improving calibration performance, which is about matching predictions regarding a random event to the long-term occurrence of the event (Dawid, 1982), is of central interest in the machine learning community due to its significance to safe and trustworthy deployment of machine learning models in critical real-world decision-making systems (Amodei et al., 2016; Lee and See, 2004). In independent and identically distributed (i.i.d.) settings, calibration performance has been significantly improved by various approaches (Gal and Ghahramani, 2016; Guo et al., 2017; Lakshminarayanan et al., 2017). However, producing well-calibrated predictions in UDA remains challenging due to the distribution shifts. Specifically, Wang et al. (2020) show the discernible compromise in calibration performance as an offset against the enhancement of target accuracy. A further observation reveals that state-of-the-art calibrated classifiers in the i.i.d.

A recently popular approach to solving reinforcement learning is with data from human preferences. In fact, human preference data are now used with classic reinforcement learning algorithms such as actor-critic methods, which involve evaluating an intermediate policy over a reward learned from human preference data with distribution shift, known as off-policy evaluation (OPE). Such algorithm includes (i) learning reward function from human preference dataset, and (ii) learning expected cumulative reward of a target policy. Despite the huge empirical success, existing OPE methods with preference data often lack theoretical understanding and rely heavily on heuristics. In this paper, we study the sample efficiency of OPE with human preference and establish a statistical guarantee for it. Specifically, we approach OPE by learning the value function by fitted-Q-evaluation with a deep neural network. By appropriately selecting the size of a ReLU network, we show that one can leverage any low-dimensional manifold structure in the Markov decision process and obtain a sample-efficient estimator without suffering from the curse of high data ambient dimensionality. Under the assumption of high reward smoothness, our results \textit{almost align with the classical OPE results with observable reward data}. To the best of our knowledge, this is the first result that establishes a \textit{provably efficient} guarantee for off-policy evaluation with RLHF.

Meta-learning has demonstrated promising results in few-shot classification (FSC) by learning to solve new problems using prior knowledge. Bayesian methods are effective at characterizing uncertainty in FSC, which is crucial in high-risk fields. In this context, the logistic-softmax likelihood is often employed as an alternative to the softmax likelihood in multi-class Gaussian process classification due to its conditional conjugacy property. However, the theoretical property of logistic-softmax is not clear and previous research indicated that the inherent uncertainty of logistic-softmax leads to suboptimal performance. To mitigate these issues, we revisit and redesign the logistic-softmax likelihood, which enables control of the \textit{a priori} confidence level through a temperature parameter. Furthermore, we theoretically and empirically show that softmax can be viewed as a special case of logistic-softmax and logistic-softmax induces a larger family of data distribution than softmax. Utilizing modified logistic-softmax, we integrate the data augmentation technique into the deep kernel based Gaussian process meta-learning framework, and derive an analytical mean-field approximation for task-specific updates. Our approach yields well-calibrated uncertainty estimates and achieves comparable or superior results on standard benchmark datasets. Code is publicly available at \url{https://github.com/keanson/revisit-logistic-softmax}.

In light of climate change's impacts on forests, including extreme drought and late-frost, leading to vitality decline and regional forest die-back, we assess univariate drought and late-frost risks and perform a joint risk analysis in Bavaria, Germany, from 1952 to 2020. Utilizing a vast dataset with 26 bioclimatic and topographic variables, we employ vine copula models due to the data's non-Gaussian and asymmetric dependencies. We use D-vine regression for univariate and Y-vine regression for bivariate analysis, and propose corresponding univariate and bivariate conditional probability risk measures. We identify "at-risk" regions, emphasizing the need for forest adaptation due to climate change.

Mutual information quantifies the dependence between two random variables and remains invariant under diffeomorphisms. In this paper, we explore the pointwise mutual information profile, an extension of mutual information that maintains this invariance. We analytically describe the profiles of multivariate normal distributions and introduce the family of fine distributions, for which the profile can be accurately approximated using Monte Carlo methods. We then show how fine distributions can be used to study the limitations of existing mutual information estimators, investigate the behavior of neural critics used in variational estimators, and understand the effect of experimental outliers on mutual information estimation. Finally, we show how fine distributions can be used to obtain model-based Bayesian estimates of mutual information, suitable for problems with available domain expertise in which uncertainty quantification is necessary.