Motion prediction systems aim to capture the future behavior of traffic scenarios enabling autonomous vehicles to perform safe and efficient planning. The evolution of these scenarios is highly uncertain and depends on the interactions of agents with static and dynamic objects in the scene. GNN-based approaches have recently gained attention as they are well suited to naturally model these interactions. However, one of the main challenges that remains unexplored is how to address the complexity and opacity of these models in order to deal with the transparency requirements for autonomous driving systems, which includes aspects such as interpretability and explainability. In this work, we aim to improve the explainability of motion prediction systems by using different approaches. First, we propose a new Explainable Heterogeneous Graph-based Policy (XHGP) model based on an heterograph representation of the traffic scene and lane-graph traversals, which learns interaction behaviors using object-level and type-level attention. This learned attention provides information about the most important agents and interactions in the scene. Second, we explore this same idea with the explanations provided by GNNExplainer. Third, we apply counterfactual reasoning to provide explanations of selected individual scenarios by exploring the sensitivity of the trained model to changes made to the input data, i.e., masking some elements of the scene, modifying trajectories, and adding or removing dynamic agents. The explainability analysis provided in this paper is a first step towards more transparent and reliable motion prediction systems, important from the perspective of the user, developers and regulatory agencies. UTONOMOUS vehicles (AVs) have to perform trajectory planning based on the global route and the local context. Trajectory planning can be applied in a safer and more efficient way if the system is able to anticipate future motions of surrounding agents [1], as humans inherently do. Motion prediction has recently gained significant attention within the research community since it is one of the key unsolved challenges in reaching full self-driving autonomy [2]. The main goal of motion prediction is to determine a set of coordinates at a future point in time for an agent in the scene. Among the different approaches, graphs are gaining attention since traffic scenarios can be naturally represented as a graph.

This paper investigates how hate speech varies in systematic ways according to the identities it targets. Across multiple hate speech datasets annotated for targeted identities, we find that classifiers trained on hate speech targeting specific identity groups struggle to generalize to other targeted identities. This provides empirical evidence for differences in hate speech by target identity; we then investigate which patterns structure this variation. We find that the targeted demographic category (e.g. gender/sexuality or race/ethnicity) appears to have a greater effect on the language of hate speech than does the relative social power of the targeted identity group. We also find that words associated with hate speech targeting specific identities often relate to stereotypes, histories of oppression, current social movements, and other social contexts specific to identities. These experiments suggest the importance of considering targeted identity, as well as the social contexts associated with these identities, in automated hate speech classification.

Automated Machine Learning-based systems' integration into a wide range of tasks has expanded as a result of their performance and speed. Although there are numerous advantages to employing ML-based systems, if they are not interpretable, they should not be used in critical, high-risk applications where human lives are at risk. To address this issue, researchers and businesses have been focusing on finding ways to improve the interpretability of complex ML systems, and several such methods have been developed. Indeed, there are so many developed techniques that it is difficult for practitioners to choose the best among them for their applications, even when using evaluation metrics. As a result, the demand for a selection tool, a meta-explanation technique based on a high-quality evaluation metric, is apparent. In this paper, we present a local meta-explanation technique which builds on top of the truthfulness metric, which is a faithfulness-based metric. We demonstrate the effectiveness of both the technique and the metric by concretely defining all the concepts and through experimentation.

Federated learning (FL) has gained significant attention recently as a privacy-enhancing tool to jointly train a machine learning model by multiple participants. The prior work on FL has mostly studied how to protect label privacy during model training. However, model evaluation in FL might also lead to potential leakage of private label information. In this work, we propose an evaluation algorithm that can accurately compute the widely used AUC (area under the curve) metric when using the label differential privacy (DP) in FL. Through extensive experiments, we show our algorithms can compute accurate AUCs compared to the ground truth. The code is available at {\url{https://github.com/bytedance/fedlearner/tree/master/example/privacy/DPAUC}}.

The regulation's scope encompasses all sectors (except for military) and aims to introduce a common regulatory and legal framework for AI, ensuring that all AI systems are safe and respect existing law on fundamental rights and values. Personally, I think AI regulation and governance is very important. We've all seen the sci-fi movies where artificially intelligent robots (sorry, beings) take over the world and attempt to bring about the end of humanity as we know it, until some bruised and battered hero saves the day. While that's the worst-case scenario meant only for our screens, there are some real use-cases for AI that are actually quite scary. Think about deepfakes, for example, where AI is used to forge an image, video, or audio recording with such precision that the average human is unlikely to detect any manipulation.

There has been a surge of interest in what generative AI can do. But what does this technology really mean for the legal sector? To find out we must navigate a path between'Death of the Lawyer 2.0' hysteria and those who dismiss the whole thing as a gimmick. Artificial Lawyer looks at what this tech can really do. Generative AI (gen AI), working via Large Language Models such as OpenAI's GPT-3, can do some amazing things.

The Uber app prompted Srikanth to try again, so he waited a few minutes and took another picture. "I was worried about bookings. We have daily targets where if we complete a certain number of bookings, we get incentives," Srikanth says. "I was anxious to log in and start driving, and not waste any time." So he tried once more.

Manasi Vartak is founder and CEO of Verta, a Palo Alto-based provider of solutions for Operational AI and ML Model Management. AI continues to transform our world as companies look to win over consumers with intelligent experiences delivered in real time on smartphones, smart TVs, smart cars--smart everything. But along with new opportunities, organizations are also finding new challenges as they seek to cross the AI chasm. Here are the top six AI/ML trends that I'll be tracking in the year ahead, along with recommendations for how enterprises can stay ahead of each trend. A recent study by our company's research group, Verta Insights, found that more than two-thirds of ML practitioners expect real-time use cases to increase significantly over the next three years.

Edutopia is a free source of information, inspiration, and practical strategies for learning and teaching in preK-12 education. We are published by the George Lucas Educational Foundation, a nonprofit, nonpartisan organization. Edutopia and Lucas Education Research are trademarks or registered trademarks of the George Lucas Educational Foundation in the U.S. and other countries.

As machine translation (MT) metrics improve their correlation with human judgement every year, it is crucial to understand the limitations of such metrics at the segment level. Specifically, it is important to investigate metric behaviour when facing accuracy errors in MT because these can have dangerous consequences in certain contexts (e.g., legal, medical). We curate ACES, a translation accuracy challenge set, consisting of 68 phenomena ranging from simple perturbations at the word/character level to more complex errors based on discourse and real-world knowledge. We use ACES to evaluate a wide range of MT metrics including the submissions to the WMT 2022 metrics shared task and perform several analyses leading to general recommendations for metric developers. We recommend: a) combining metrics with different strengths, b) developing metrics that give more weight to the source and less to surface-level overlap with the reference and c) explicitly modelling additional language-specific information beyond what is available via multilingual embeddings.