We visualize the domain adaptation performance on the transfer scenario NIH-CXR14, CheXpert, MIMIC-CXR to Open-i . The visualization sample in the Open-i is suffering from Atelecsis and Effusion disease.

Vision models excel in image classification but struggle to generalize to unseen data, such as classifying images from unseen domains or discovering novel categories. In this paper, we explore the relationship between logical reasoning and deep learning generalization in visual classification. A logical regularization termed L-Reg is derived which bridges a logical analysis framework to image classification. Our work reveals that L-Reg reduces the complexity of the model in terms of the feature distribution and classifier weights. Specifically, we unveil the interpretability brought by L-Reg, as it enables the model to extract the salient features, such as faces to persons, for classification.

Time series data captures properties that change over time. Such data occurs widely, ranging from the scientific and medical domains to the industrial and environmental domains. When the properties in time series exhibit spatial variations, we often call the data spatio-temporal. As part of the continued digitalization of processes throughout society, increasingly large volumes of time series and spatio-temporal data are available. In this tutorial, we focus on data-driven decision making with such data, e.g., enabling greener and more efficient transportation based on traffic time series forecasting. The tutorial adopts the holistic paradigm of "data-governance-analytics-decision." We first introduce the data foundation of time series and spatio-temporal data, which is often heterogeneous. Next, we discuss data governance methods that aim to improve data quality. We then cover data analytics, focusing on five desired characteristics: automation, robustness, generality, explainability, and resource efficiency. We finally cover data-driven decision making strategies and briefly discuss promising research directions. We hope that the tutorial will serve as a primary resource for researchers and practitioners who are interested in value creation from time series and spatio-temporal data.

Learning curves are a concept from social sciences that has been adopted in the context of machine learning to assess the performance of a learning algorithm with respect to a certain resource, e.g., the number of training examples or the number of training iterations. Learning curves have important applications in several machine learning contexts, most notably in data acquisition, early stopping of model training, and model selection. For instance, learning curves can be used to model the performance of the combination of an algorithm and its hyperparameter configuration, providing insights into their potential suitability at an early stage and often expediting the algorithm selection process. Various learning curve models have been proposed to use learning curves for decision making. Some of these models answer the binary decision question of whether a given algorithm at a certain budget will outperform a certain reference performance, whereas more complex models predict the entire learning curve of an algorithm. We contribute a framework that categorises learning curve approaches using three criteria: the decision-making situation they address, the intrinsic learning curve question they answer and the type of resources they use. We survey papers from the literature and classify them into this framework.

The news: Self-driving car startup Wayve can now interrogate its vehicles, asking them questions about their driving decisions--and getting answers back thanks to a chatbot. How it works: The idea is to use the same tech behind ChatGPT to help train driverless cars. The company combined its existing self-driving software with a large language model, creating a hybrid model that syncs up video data and driving data with natural-language descriptions that capture what the car sees and what it does. Why it matters: Wayve is treating the news as a breakthrough in AI safety. By quizzing its self-driving software every step of the way, Wayve hopes to understand exactly why and how its cars make certain decisions--and to uncover mistakes more quickly.

I'm writing this post from the guest room in my mom's house, which is peppered with old knick-knacks of mine--to summon the spirit of my childhood room, I suppose. While flipping through my photo albums, I was tickled to find a blurry picture of the poster for Phone Booth, clearly taken by me on a disposable camera outside of a movie theater. I was probably too young to be watching a gunman thriller--thanks, Mom--but I'm pretty sure my affection for it had a lot to do with Colin Farrell, who was a relative unknown when that movie came out in 2002. To this day, I'm a bit gaga over him, though I think part of the reason my puppy love has turned into something more enduring is that, as I've gotten older and my tastes have evolved, so has the actor's persona. Not to downplay his macho heartthrob phase in the aughts--I still go catatonic whenever I think about him salsa dancing in Miami Vice, and I sense noted MV-heads Bilge and David feel the same way--but it has been a delight to see him take on increasingly stranger, more cerebral roles for directors like Yorgos Lanthimos and Sofia Coppola while also pushing himself, unafraid to get ugly and unhinged, in blockbusters like The Batman.

Marketers have tons of data at their disposal. But 80% say they have trouble making data-driven decisions, according to a study from Pecan AI, conducted by Wakefield Research. Worse, 90% of those with AI-powered predictive analytics have trouble with decision-making. Yet 95% of companies now integrate that capability into their marketing strategy, and 44% have done so completely. Why are there so many problems?

In open-retrieval conversational machine reading (OR-CMR) task, machines are required to do multi-turn question answering given dialogue history and a textual knowledge base. Existing works generally utilize two independent modules to approach this problem's two successive sub-tasks: first with a hard-label decision making and second with a question generation aided by various entailment reasoning methods. Such usual cascaded modeling is vulnerable to error propagation and prevents the two sub-tasks from being consistently optimized. In this work, we instead model OR-CMR as a unified text-to-text task in a fully end-to-end style. Experiments on the OR-ShARC dataset show the effectiveness of our proposed end-to-end framework on both sub-tasks by a large margin, achieving new state-of-the-art results. Further ablation studies support that our framework can generalize to different backbone models.

Imagine experiencing a crash as the passenger of an autonomous vehicle. Wouldn't you want to know why it happened? Current end-to-end optimizable deep neural networks (DNNs) in 3D detection, multi-object tracking, and motion forecasting provide little to no explanations about how they make their decisions. To help bridge this gap, we design an end-to-end optimizable multi-object tracking architecture and training protocol inspired by the recently proposed method of interchange intervention training (IIT). By enumerating different tracking decisions and associated reasoning procedures, we can train individual networks to reason about the possible decisions via IIT. Each network's decisions can be explained by the high-level structural causal model (SCM) it is trained in alignment with. Moreover, our proposed model learns to rank these outcomes, leveraging the promise of deep learning in end-to-end training, while being inherently interpretable.