The alfalfa crop is globally important as livestock feed, so highly efficient planting and harvesting could benefit many industries, especially as the global climate changes and traditional methods become less accurate. Recent work using machine learning (ML) to predict yields for alfalfa and other crops has shown promise. Previous efforts used remote sensing, weather, planting, and soil data to train machine learning models for yield prediction. However, while remote sensing works well, the models require large amounts of data and cannot make predictions until the harvesting season begins. Using weather and planting data from alfalfa variety trials in Kentucky and Georgia, our previous work compared feature selection techniques to find the best technique and best feature set. In this work, we trained a variety of machine learning models, using cross validation for hyperparameter optimization, to predict biomass yields, and we showed better accuracy than similar work that employed more complex techniques. Our best individual model was a random forest with a mean absolute error of 0.081 tons/acre and R{$^2$} of 0.941. Next, we expanded this dataset to include Wisconsin and Mississippi, and we repeated our experiments, obtaining a higher best R{$^2$} of 0.982 with a regression tree. We then isolated our testing datasets by state to explore this problem's eligibility for domain adaptation (DA), as we trained on multiple source states and tested on one target state. This Trivial DA (TDA) approach leaves plenty of room for improvement through exploring more complex DA techniques in forthcoming work.

Complex networks representing social interactions, brain activities, molecular structures have been studied widely to be able to understand and predict their characteristics as graphs. Models and algorithms for these networks are used in real-life applications, such as search engines, and recommender systems. In general, such networks are modelled by constructing a low-dimensional Euclidean embedding of the vertices of the network, where proximity of the vertices in the Euclidean space hints the likelihood of an edge (link). In this work, we study the performance of such low-rank representations of real-life networks on a network classification problem.

This paper finds that contrastive learning can produce superior sentence embeddings for pre-trained models but is also vulnerable to backdoor attacks. We present the first backdoor attack framework, BadCSE, for state-of-the-art sentence embeddings under supervised and unsupervised learning settings. The attack manipulates the construction of positive and negative pairs so that the backdoored samples have a similar embedding with the target sample (targeted attack) or the negative embedding of its clean version (non-targeted attack). By injecting the backdoor in sentence embeddings, BadCSE is resistant against downstream fine-tuning. We evaluate BadCSE on both STS tasks and other downstream tasks. The supervised non-targeted attack obtains a performance degradation of 194.86%, and the targeted attack maps the backdoored samples to the target embedding with a 97.70% success rate while maintaining the model utility.

Accurate co-estimations of battery states, such as state-of-charge (SOC), state-of-health (SOH,) and remaining useful life (RUL), are crucial to the battery management systems to assure safe and reliable management. Although the external properties of the battery charge with the aging degree, batteries' degradation mechanism shares similar evolving patterns. Since batteries are complicated chemical systems, these states are highly coupled with intricate electrochemical processes. A state-coupled co-estimation method named Deep Inter and Intra-Cycle Attention Network (DIICAN) is proposed in this paper to estimate SOC, SOH, and RUL, which organizes battery measurement data into the intra-cycle and inter-cycle time scales. And to extract degradation-related features automatically and adapt to practical working conditions, the convolutional neural network is applied. The state degradation attention unit is utilized to extract the battery state evolution pattern and evaluate the battery degradation degree. To account for the influence of battery aging on the SOC estimation, the battery degradation-related state is incorporated in the SOC estimation for capacity calibration. The DIICAN method is validated on the Oxford battery dataset. The experimental results show that the proposed method can achieve SOH and RUL co-estimation with high accuracy and effectively improve SOC estimation accuracy for the whole lifespan.

This paper considers the Pointer Value Retrieval (PVR) benchmark introduced in [ZRKB21], where a 'reasoning' function acts on a string of digits to produce the label. More generally, the paper considers the learning of logical functions with gradient descent (GD) on neural networks. It is first shown that in order to learn logical functions with gradient descent on symmetric neural networks, the generalization error can be lower-bounded in terms of the noise-stability of the target function, supporting a conjecture made in [ZRKB21]. It is then shown that in the distribution shift setting, when the data withholding corresponds to freezing a single feature (referred to as canonical holdout), the generalization error of gradient descent admits a tight characterization in terms of the Boolean influence for several relevant architectures. This is shown on linear models and supported experimentally on other models such as MLPs and Transformers. In particular, this puts forward the hypothesis that for such architectures and for learning logical functions such as PVR functions, GD tends to have an implicit bias towards low-degree representations, which in turn gives the Boolean influence for the generalization error under quadratic loss.

With the rapid development of machine learning, improving its explainability has become a crucial research goal. We study the problem of making the clusters more explainable by investigating the cluster descriptors. Given a set of objects $S$, a clustering of these objects $\pi$, and a set of tags $T$ that have not participated in the clustering algorithm. Each object in $S$ is associated with a subset of $T$. The goal is to find a representative set of tags for each cluster, referred to as the cluster descriptors, with the constraint that these descriptors we find are pairwise disjoint, and the total size of all the descriptors is minimized. In general, this problem is NP-hard. We propose a novel explainability model that reinforces the previous models in such a way that tags that do not contribute to explainability and do not sufficiently distinguish between clusters are not added to the optimal descriptors. The proposed model is formulated as a quadratic unconstrained binary optimization problem which makes it suitable for solving on modern optimization hardware accelerators. We experimentally demonstrate how a proposed explainability model can be solved on specialized hardware for accelerating combinatorial optimization, the Fujitsu Digital Annealer, and use real-life Twitter and PubMed datasets for use cases.

In fault detection and diagnosis of prognostics and health management (PHM) systems, most of the methodologies utilize machine learning (ML) or deep learning (DL) through which either some features are extracted beforehand (in the case of ML) or filters are used to extract features autonomously (in case of DL) to perform the critical classification task. Particularly in the fault detection and diagnosis of industrial robots where electric current, vibration or acoustic emissions signals are the primary sources of information, a feature domain that can map the signals into their constituent components with compressed information at different levels can reduce the complexities and size of typical ML and DL-based frameworks. The Deep Scattering Spectrum (DSS) is one of the strategies that use the Wavelet Transform (WT) analogy to separate and extract the information encoded in a signal's various temporal and frequency domains. As a result, the focus of this work is on the study of the DSS's relevance to fault detection and daignosis for mechanical components of industrail robots. We used multiple industrial robots and distinct mechanical faults to build an approach for classifying the faults using low-variance features extracted from the input signals. The presented approach was implemented on the practical test benches and demonstrated satisfactory performance in fault detection and diagnosis for simple and complex classification problems with a classification accuracy of 99.7% and 88.1%, respectively.

A cognitive radar is a constrained utility maximizer that adapts its sensing mode in response to a changing environment. If an adversary can estimate the utility function of a cognitive radar, it can determine the radar's sensing strategy and mitigate the radar performance via electronic countermeasures (ECM). This paper discusses how a cognitive radar can {\em hide} its strategy from an adversary that detects cognition. The radar does so by transmitting purposefully designed sub-optimal responses to spoof the adversary's Neyman-Pearson detector. We provide theoretical guarantees by ensuring the Type-I error probability of the adversary's detector exceeds a pre-defined level for a specified tolerance on the radar's performance loss. We illustrate our cognition masking scheme via numerical examples involving waveform adaptation and beam allocation. We show that small purposeful deviations from the optimal strategy of the radar confuse the adversary by significant amounts, thereby masking the radar's cognition. Our approach uses novel ideas from revealed preference in microeconomics and adversarial inverse reinforcement learning. Our proposed algorithms provide a principled approach for system-level electronic counter-countermeasures (ECCM) to mask the radar's cognition, i.e., hide the radar's strategy from an adversary. We also provide performance bounds for our cognition masking scheme when the adversary has misspecified measurements of the radar's response.

Deep classifiers are known to rely on spurious features $\unicode{x2013}$ patterns which are correlated with the target on the training data but not inherently relevant to the learning problem, such as the image backgrounds when classifying the foregrounds. In this paper we evaluate the amount of information about the core (non-spurious) features that can be decoded from the representations learned by standard empirical risk minimization (ERM) and specialized group robustness training. Following recent work on Deep Feature Reweighting (DFR), we evaluate the feature representations by re-training the last layer of the model on a held-out set where the spurious correlation is broken. On multiple vision and NLP problems, we show that the features learned by simple ERM are highly competitive with the features learned by specialized group robustness methods targeted at reducing the effect of spurious correlations. Moreover, we show that the quality of learned feature representations is greatly affected by the design decisions beyond the training method, such as the model architecture and pre-training strategy. On the other hand, we find that strong regularization is not necessary for learning high quality feature representations. Finally, using insights from our analysis, we significantly improve upon the best results reported in the literature on the popular Waterbirds, CelebA hair color prediction and WILDS-FMOW problems, achieving 97%, 92% and 50% worst-group accuracies, respectively.

In performative prediction, a predictive model impacts the distribution that generates future data, a phenomenon that is being ignored in classical supervised learning. In this closed-loop setting, the natural measure of performance named performative risk ($\mathrm{PR}$), captures the expected loss incurred by a predictive model \emph{after} deployment. The core difficulty of using the performative risk as an optimization objective is that the data distribution itself depends on the model parameters. This dependence is governed by the environment and not under the control of the learner. As a consequence, even the choice of a convex loss function can result in a highly non-convex $\mathrm{PR}$ minimization problem. Prior work has identified a pair of general conditions on the loss and the mapping from model parameters to distributions that implies the convexity of the performative risk. In this paper, we relax these assumptions and focus on obtaining weaker notions of convexity, without sacrificing the amenability of the $\mathrm{PR}$ minimization problem for iterative optimization methods.