Robots need to be able to adapt to unexpected changes in the environment such that they can autonomously succeed in their tasks. However, hand-designing feedback models for adaptation is tedious, if at all possible, making data-driven methods a promising alternative. In this paper we introduce a full framework for learning feedback models for reactive motion planning. Our pipeline starts by segmenting demonstrations of a complete task into motion primitives via a semi-automated segmentation algorithm. Then, given additional demonstrations of successful adaptation behaviors, we learn initial feedback models through learning from demonstrations. In the final phase, a sample-efficient reinforcement learning algorithm fine-tunes these feedback models for novel task settings through few real system interactions. We evaluate our approach on a real anthropomorphic robot in learning a tactile feedback task.

We present an effective method for fusing visual-and-language representations for several question answering tasks including visual question answering and visual entailment. In contrast to prior works that concatenate unimodal representations or use only cross-attention, we compose multimodal representations via channel fusion. By fusing on the channels, the model is able to more effectively align the tokens compared to standard methods. These multimodal representations, which we call compound tokens are generated with cross-attention transformer layers. First, vision tokens are used as queries to retrieve compatible text tokens through cross-attention. We then chain the vision tokens and the queried text tokens along the channel dimension. We call the resulting representations compound tokens. A second group of compound tokens are generated using an analogous process where the text tokens serve as queries to the cross-attention layer. We demonstrate the effectiveness of compound tokens using an encoder-decoder vision-language model trained end-to-end in the open-vocabulary setting. Compound Tokens achieve highly competitive performance across a range of question answering tasks including GQA, VQA2.0, and SNLI-VE.

In this paper we generalize single-relation pseudo-Riemannian graph embedding models to multi-relational networks, and show that the typical approach of encoding relations as manifold transformations translates from the Riemannian to the pseudo-Riemannian case. In addition we construct a view of relations as separate spacetime submanifolds of multi-time manifolds, and consider an interpolation between a pseudo-Riemannian embedding model and its Wick-rotated Riemannian counterpart. We validate these extensions in the task of link prediction, focusing on flat Lorentzian manifolds, and demonstrate their use in both knowledge graph completion and knowledge discovery in a biological domain.

We introduce an ensemble learning method based on Gaussian Process Regression (GPR) for predicting conditional expected stock returns given stock-level and macro-economic information. Our ensemble learning approach significantly reduces the computational complexity inherent in GPR inference and lends itself to general online learning tasks. We conduct an empirical analysis on a large cross-section of US stocks from 1962 to 2016. We find that our method dominates existing machine learning models statistically and economically in terms of out-of-sample $R$-squared and Sharpe ratio of prediction-sorted portfolios. Exploiting the Bayesian nature of GPR, we introduce the mean-variance optimal portfolio with respect to the predictive uncertainty distribution of the expected stock returns. It appeals to an uncertainty averse investor and significantly dominates the equal- and value-weighted prediction-sorted portfolios, which outperform the S&P 500.

Different machine learning (ML) models are trained on SCADA and meteorological data collected at an onshore wind farm and then assessed in terms of fidelity and accuracy for predictions of wind speed, turbulence intensity, and power capture at the turbine and wind farm levels for different wind and atmospheric conditions. ML methods for data quality control and pre-processing are applied to the data set under investigation and found to outperform standard statistical methods. A hybrid model, comprised of a linear interpolation model, Gaussian process, deep neural network (DNN), and support vector machine, paired with a DNN filter, is found to achieve high accuracy for modeling wind turbine power capture. Modifications of the incoming freestream wind speed and turbulence intensity, $TI$, due to the evolution of the wind field over the wind farm and effects associated with operating turbines are also captured using DNN models. Thus, turbine-level modeling is achieved using models for predicting power capture while farm-level modeling is achieved by combining models predicting wind speed and $TI$ at each turbine location from freestream conditions with models predicting power capture. Combining these models provides results consistent with expected power capture performance and holds promise for future endeavors in wind farm modeling and diagnostics. Though training ML models is computationally expensive, using the trained models to simulate the entire wind farm takes only a few seconds on a typical modern laptop computer, and the total computational cost is still lower than other available mid-fidelity simulation approaches.

Market prediction plays a major role in supporting financial decisions. An emerging approach in this domain is to use graphical modeling and analysis to for prediction of next market index fluctuations. One important question in this domain is how to construct an appropriate graphical model of the data that can be effectively used by a semi-supervised GNN to predict index fluctuations. In this paper, we introduce a framework called NETpred that generates a novel heterogeneous graph representing multiple related indices and their stocks by using several stock-stock and stock-index relation measures. It then thoroughly selects a diverse set of representative nodes that cover different parts of the state space and whose price movements are accurately predictable. By assigning initial predicted labels to such a set of nodes, NETpred makes sure that the subsequent GCN model can be successfully trained using a semi-supervised learning process. The resulting model is then used to predict the stock labels which are finally aggregated to infer the labels for all the index nodes in the graph. Our comprehensive set of experiments shows that NETpred improves the performance of the state-of-the-art baselines by 3%-5% in terms of F-score measure on different well-known data sets.

Eye movements are known to reflect cognitive processes in reading, and psychological reading research has shown that eye gaze patterns differ between readers with and without dyslexia. In recent years, researchers have attempted to classify readers with dyslexia based on their eye movements using Support Vector Machines (SVMs). However, these approaches (i) are based on highly aggregated features averaged over all words read by a participant, thus disregarding the sequential nature of the eye movements, and (ii) do not consider the linguistic stimulus and its interaction with the reader's eye movements. In the present work, we propose two simple sequence models that process eye movements on the entire stimulus without the need of aggregating features across the sentence. Additionally, we incorporate the linguistic stimulus into the model in two ways -- contextualized word embeddings and manually extracted linguistic features. The models are evaluated on a Mandarin Chinese dataset containing eye movements from children with and without dyslexia. Our results show that (i) even for a logographic script such as Chinese, sequence models are able to classify dyslexia on eye gaze sequences, reaching state-of-the-art performance, and (ii) incorporating the linguistic stimulus does not help to improve classification performance.

Online learning naturally arises in many statistical and machine learning problems. The most widely used methods in online learning are stochastic first-order algorithms. Among this family of algorithms, there is a recently developed algorithm, Recursive One-Over-T SGD (ROOT-SGD). ROOT-SGD is advantageous in that it converges at a non-asymptotically fast rate, and its estimator further converges to a normal distribution. However, this normal distribution has unknown asymptotic covariance; thus cannot be directly applied to measure the uncertainty. To fill this gap, we develop two estimators for the asymptotic covariance of ROOT-SGD. Our covariance estimators are useful for statistical inference in ROOT-SGD. Our first estimator adopts the idea of plug-in. For each unknown component in the formula of the asymptotic covariance, we substitute it with its empirical counterpart. The plug-in estimator converges at the rate $\mathcal{O}(1/\sqrt{t})$, where $t$ is the sample size. Despite its quick convergence, the plug-in estimator has the limitation that it relies on the Hessian of the loss function, which might be unavailable in some cases. Our second estimator is a Hessian-free estimator that overcomes the aforementioned limitation. The Hessian-free estimator uses the random-scaling technique, and we show that it is an asymptotically consistent estimator of the true covariance.

The problem of covariate-shift generalization has attracted intensive research attention. Previous stable learning algorithms employ sample reweighting schemes to decorrelate the covariates when there is no explicit domain information about training data. However, with finite samples, it is difficult to achieve the desirable weights that ensure perfect independence to get rid of the unstable variables. Besides, decorrelating within stable variables may bring about high variance of learned models because of the over-reduced effective sample size. A tremendous sample size is required for these algorithms to work. In this paper, with theoretical justification, we propose SVI (Sparse Variable Independence) for the covariate-shift generalization problem. We introduce sparsity constraint to compensate for the imperfectness of sample reweighting under the finite-sample setting in previous methods. Furthermore, we organically combine independence-based sample reweighting and sparsity-based variable selection in an iterative way to avoid decorrelating within stable variables, increasing the effective sample size to alleviate variance inflation. Experiments on both synthetic and real-world datasets demonstrate the improvement of covariate-shift generalization performance brought by SVI.

The space-air-ground integrated network (SAGIN), one of the key technologies for next-generation mobile communication systems, can facilitate data transmission for users all over the world, especially in some remote areas where vast amounts of informative data are collected by Internet of remote things (IoRT) devices to support various data-driven artificial intelligence (AI) services. However, training AI models centrally with the assistance of SAGIN faces the challenges of highly constrained network topology, inefficient data transmission, and privacy issues. To tackle these challenges, we first propose a novel topology-aware federated learning framework for the SAGIN, namely Olive Branch Learning (OBL). Specifically, the IoRT devices in the ground layer leverage their private data to perform model training locally, while the air nodes in the air layer and the ring-structured low earth orbit (LEO) satellite constellation in the space layer are in charge of model aggregation (synchronization) at different scales.To further enhance communication efficiency and inference performance of OBL, an efficient Communication and Non-IID-aware Air node-Satellite Assignment (CNASA) algorithm is designed by taking the data class distribution of the air nodes as well as their geographic locations into account. Furthermore, we extend our OBL framework and CNASA algorithm to adapt to more complex multi-orbit satellite networks. We analyze the convergence of our OBL framework and conclude that the CNASA algorithm contributes to the fast convergence of the global model. Extensive experiments based on realistic datasets corroborate the superior performance of our algorithm over the benchmark policies.