Seizure type identification is essential for the treatment and management of epileptic patients. However, it is a difficult process known to be time consuming and labor intensive. Automated diagnosis systems, with the advancement of machine learning algorithms, have the potential to accelerate the classification process, alert patients, and support physicians in making quick and accurate decisions. In this paper, we present a novel multi-path seizure-type classification deep learning network (MP-SeizNet), consisting of a convolutional neural network (CNN) and a bidirectional long short-term memory neural network (Bi-LSTM) with an attention mechanism. The objective of this study was to classify specific types of seizures, including complex partial, simple partial, absence, tonic, and tonic-clonic seizures, using only electroencephalogram (EEG) data. The EEG data is fed to our proposed model in two different representations. The CNN was fed with wavelet-based features extracted from the EEG signals, while the Bi-LSTM was fed with raw EEG signals to let our MP-SeizNet jointly learns from different representations of seizure data for more accurate information learning. The proposed MP-SeizNet was evaluated using the largest available EEG epilepsy database, the Temple University Hospital EEG Seizure Corpus, TUSZ v1.5.2. We evaluated our proposed model across different patient data using three-fold cross-validation and across seizure data using five-fold cross-validation, achieving F1 scores of 87.6% and 98.1%, respectively.

The transfer learning paradigm of model pre-training and subsequent fine-tuning produces high-accuracy models. While most studies recommend scaling the pre-training size to benefit most from transfer learning, a question remains: what data and method should be used for pre-training? We investigate the impact of pre-training data distribution on the few-shot and full fine-tuning performance using 3 pre-training methods (supervised, contrastive language-image and image-image), 7 pre-training datasets, and 9 downstream datasets. Through extensive controlled experiments, we find that the choice of the pre-training data source is essential for the few-shot transfer, but its role decreases as more data is made available for fine-tuning. Additionally, we explore the role of data curation and examine the trade-offs between label noise and the size of the pre-training dataset. We find that using 2000X more pre-training data from LAION can match the performance of supervised ImageNet pre-training. Furthermore, we investigate the effect of pre-training methods, comparing language-image contrastive vs. image-image contrastive, and find that the latter leads to better downstream accuracy

Climate change has increased the severity and frequency of weather disasters all around the world. Flood inundation mapping based on earth observation data can help in this context, by providing cheap and accurate maps depicting the area affected by a flood event to emergency-relief units in near-real-time. Building upon the recent development of the Sen1Floods11 dataset, which provides a limited amount of hand-labeled high-quality training data, this paper evaluates the potential of five traditional machine learning approaches such as gradient boosted decision trees, support vector machines or quadratic discriminant analysis. By performing a grid-search-based hyperparameter optimization on 23 feature spaces we can show that all considered classifiers are capable of outperforming the current state-of-the-art neural network-based approaches in terms of total IoU on their best-performing feature spaces. With total and mean IoU values of 0.8751 and 0.7031 compared to 0.70 and 0.5873 as the previous best-reported results, we show that a simple gradient boosting classifier can significantly improve over deep neural network based approaches, despite using less training data. Furthermore, an analysis of the regional distribution of the Sen1Floods11 dataset reveals a problem of spatial imbalance. We show that traditional machine learning models can learn this bias and argue that modified metric evaluations are required to counter artifacts due to spatial imbalance. Lastly, a qualitative analysis shows that this pixel-wise classifier provides highly-precise surface water classifications indicating that a good choice of a feature space and pixel-wise classification can generate high-quality flood maps using optical and SAR data. We make our code publicly available at: https://github.com/DFKI-Earth-And-Space-Applications/Flood_Mapping_Feature_Space_Importance

A big convergence of language, multimodal perception, action, and world modeling is a key step toward artificial general intelligence. In this work, we introduce Kosmos-1, a Multimodal Large Language Model (MLLM) that can perceive general modalities, learn in context (i.e., few-shot), and follow instructions (i.e., zero-shot). Specifically, we train Kosmos-1 from scratch on web-scale multimodal corpora, including arbitrarily interleaved text and images, image-caption pairs, and text data. We evaluate various settings, including zero-shot, few-shot, and multimodal chain-of-thought prompting, on a wide range of tasks without any gradient updates or finetuning. Experimental results show that Kosmos-1 achieves impressive performance on (i) language understanding, generation, and even OCR-free NLP (directly fed with document images), (ii) perception-language tasks, including multimodal dialogue, image captioning, visual question answering, and (iii) vision tasks, such as image recognition with descriptions (specifying classification via text instructions). We also show that MLLMs can benefit from cross-modal transfer, i.e., transfer knowledge from language to multimodal, and from multimodal to language. In addition, we introduce a dataset of Raven IQ test, which diagnoses the nonverbal reasoning capability of MLLMs.

Aspect term extraction is a fundamental task in fine-grained sentiment analysis, which aims at detecting customer's opinion targets from reviews on product or service. The traditional supervised models can achieve promising results with annotated datasets, however, the performance dramatically decreases when they are applied to the task of cross-domain aspect term extraction. Existing cross-domain transfer learning methods either directly inject linguistic features into Language models, making it difficult to transfer linguistic knowledge to target domain, or rely on the fixed predefined prompts, which is time-consuming to construct the prompts over all potential aspect term spans. To resolve the limitations, we propose a soft prompt-based joint learning method for cross domain aspect term extraction in this paper. Specifically, by incorporating external linguistic features, the proposed method learn domain-invariant representations between source and target domains via multiple objectives, which bridges the gap between domains with varied distributions of aspect terms. Further, the proposed method interpolates a set of transferable soft prompts consisted of multiple learnable vectors that are beneficial to detect aspect terms in target domain. Extensive experiments are conducted on the benchmark datasets and the experimental results demonstrate the effectiveness of the proposed method for cross-domain aspect terms extraction.

Recent methods for imitation learning directly learn a Q-function using an implicit reward formulation rather than an explicit reward function. However, these methods generally require implicit reward regularization to improve stability and often mistreat absorbing states. Previous works show that a squared norm regularization on the implicit reward function is effective, but do not provide a theoretical analysis of the resulting properties of the algorithms. This perspective allows us to address instabilities and properly treat absorbing states. We show that our method, Least Squares Inverse Q-Learning (LS-IQ), outperforms state-of-the-art algorithms, particularly in environments with absorbing states. Finally, we propose to use an inverse dynamics model to learn from observations only. Using this approach, we retain performance in settings where no expert actions are available. Inverse Reinforcement Learning (IRL) techniques have been developed to robustly extract behaviors from expert demonstration and solve the problems of classical Imitation Learning (IL) methods (Ng et al., 1999; Ziebart et al., 2008). Among the recent methods for IRL, the Adversarial Imitation Learning (AIL) approach (Ho & Ermon, 2016; Fu et al., 2018; Peng et al., 2021), which casts the optimization over rewards and policies into an adversarial setting, have been proven particularly successful. These methods, inspired by Generative Adversarial Networks (GANs) (Goodfellow et al., 2014), alternate between learning a discriminator, and improving the agent's policy w.r.t. a reward function, computed based on the discriminator's output. These explicit reward methods require many interactions with the environment as they learn both a reward and a value function.

The GPT-3.5 models have demonstrated impressive performance in various Natural Language Processing (NLP) tasks, showcasing their strong understanding and reasoning capabilities. However, their robustness and abilities to handle various complexities of the open world have yet to be explored, which is especially crucial in assessing the stability of models and is a key aspect of trustworthy AI. In this study, we perform a comprehensive experimental analysis of GPT-3.5, exploring its robustness using 21 datasets (about 116K test samples) with 66 text transformations from TextFlint that cover 9 popular Natural Language Understanding (NLU) tasks. Our findings indicate that while GPT-3.5 outperforms existing fine-tuned models on some tasks, it still encounters significant robustness degradation, such as its average performance dropping by up to 35.74\% and 43.59\% in natural language inference and sentiment analysis tasks, respectively. We also show that GPT-3.5 faces some specific robustness challenges, including robustness instability, prompt sensitivity, and number sensitivity. These insights are valuable for understanding its limitations and guiding future research in addressing these challenges to enhance GPT-3.5's overall performance and generalization abilities.

A central problem in multivariate forecasting is the need to forecast a large group of time series arranged in a natural hierarchical structure, such that time series at higher levels of the hierarchy are aggregates of time series at lower levels. For example, hierarchical time series are common in retail forecasting applications [Fildes et al., 2019], where the time series may capture retail sales of a company at different granularities such as item-level sales, category-level sales, and department-level sales. In electricity demand forecasting [Van Erven and Cugliari, 2015], the time series may correspond to electricity consumption at different granularities, starting with individual households, which could be progressively grouped into city-level, and then state-level consumption time-series. The hierarchical structure among the time series is usually represented as a tree, with leaf-level nodes corresponding to time series at the finest granularity, while higher-level nodes represent coarser-granularities and are obtained by aggregating the values from its children nodes. Since businesses usually require forecasts at various different granularities, the goal is to obtain accurate forecasts for time series at every level of the hierarchy. Furthermore, to ensure decisionmaking at different hierarchical levels are aligned, it is essential to generate predictions that are coherent [Hyndman et al., 2011] with respect to the hierarchy, that is, the forecasts of a parent time-series should be equal to the sum of forecasts of its children time-series.

We tackle the problem of Selective Classification where the objective is to achieve the best performance on a predetermined ratio (coverage) of the dataset. Recent state-of-the-art selective methods come with architectural changes either via introducing a separate selection head or an extra abstention logit. In this paper, we challenge the aforementioned methods. The results suggest that the superior performance of state-of-the-art methods is owed to training a more generalizable classifier rather than their proposed selection mechanisms. We argue that the best performing selection mechanism should instead be rooted in the classifier itself. Our proposed selection strategy uses the classification scores and achieves better results by a significant margin, consistently, across all coverages and all datasets, without any added compute cost. Furthermore, inspired by semi-supervised learning, we propose an entropy-based regularizer that improves the performance of selective classification methods. Our proposed selection mechanism with the proposed entropy-based regularizer achieves new state-of-the-art results. A model's ability to abstain from a decision when lacking confidence is essential in mission-critical applications. This is known as the Selective Prediction problem setting. The abstained and uncertain samples can be flagged and passed to a human expert for manual assessment, which, in turn, can improve the re-training process. This is crucial in problem settings where confidence is critical or an incorrect prediction can have significant consequences such as in the financial, medical, or autonomous driving domains. Several papers have tried to address this problem by estimating the uncertainty in the prediction.

Generating accurate digital tree models from scanned environments is invaluable for forestry, agriculture, and other outdoor industries in tasks such as identifying biomass, fall hazards and traversability, as well as digital applications such as animation and gaming. Existing methods for tree reconstruction rely on feature identification (trunk, crown, etc) to heuristically segment a forest into individual trees and generate a branch structure graph, limiting their application to sparse trees and uniform forests. However, the natural world is a messy place in which trees present with significant heterogeneity and are frequently encroached upon by the surrounding environment. We present a general method for extracting the branch structure of trees from point cloud data, which estimates the structure of trees by adapting the methods of structural topology optimisation to find the optimal material distribution to support wind-loading. We present the results of this optimisation over a wide variety of scans, and discuss the benefits and drawbacks of this novel approach to tree structure reconstruction. Despite the high variability of datasets containing trees, and the high rate of occlusions, our method generates detailed and accurate tree structures in most cases.