Right now, you can grab an Amazon Echo Show 5 bundle for just 46.98, which amounts to the Echo Show 5 at 50 percent off along with a near-free 60W A19 smart bulb that's controllable via the Echo Show. The beauty of the Echo Show 5 is that it's a fairly inexpensive way to bring a smart screen assistant into your home. With a 5-inch display that shows everything from the current time and date to your active Spotify playlist to the results to your voice queries and even your smart doorbell's live video feed, the Echo Show serves perfectly anywhere in your home. You can ask Alexa to control the light bulb that comes with it, play music, read out recipes, turn off your compatible smart TV, and more. If you dive in deeper, you can even create custom Alexa routines so when you say something like "Alexa, good night," it responds by turning off the lights and the TV for you.

People have a variety of preferences for how robots behave. To understand and reason about these preferences, robots aim to learn a reward function that describes how aligned robot behaviors are with a user's preferences. Good representations of a robot's behavior can significantly reduce the time and effort required for a user to teach the robot their preferences. Specifying these representations -- what "features" of the robot's behavior matter to users -- remains a difficult problem; Features learned from raw data lack semantic meaning and features learned from user data require users to engage in tedious labeling processes. Our key insight is that users tasked with customizing a robot are intrinsically motivated to produce labels through exploratory search; they explore behaviors that they find interesting and ignore behaviors that are irrelevant. To harness this novel data source of exploratory actions, we propose contrastive learning from exploratory actions (CLEA) to learn trajectory features that are aligned with features that users care about. We learned CLEA features from exploratory actions users performed in an open-ended signal design activity (N=25) with a Kuri robot, and evaluated CLEA features through a second user study with a different set of users (N=42). CLEA features outperformed self-supervised features when eliciting user preferences over four metrics: completeness, simplicity, minimality, and explainability.

Transformer based models are increasingly being used in various domains including recommender systems (RS). Pretrained transformer models such as BERT have shown good performance at language modelling. With the greater ability to model sequential tasks, variants of Encoder-only models (like BERT4Rec, SASRec etc.) have found success in sequential RS problems. Computing dot-product attention in traditional transformer models has quadratic complexity in sequence length. This is a bigger problem with RS because unlike language models, new items are added to the catalogue every day. User buying history is a dynamic sequence which depends on multiple factors. Recently, various linear attention models have tried to solve this problem by making the model linear in sequence length (token dimensions). Hydra attention is one such linear complexity model proposed for vision transformers which reduces the complexity of attention for both the number of tokens as well as model embedding dimensions. Building on the idea of Hydra attention, we introduce an efficient Transformer based Sequential RS (HydraRec) which significantly improves theoretical complexity of computing attention for longer sequences and bigger datasets while preserving the temporal context. Extensive experiments are conducted to evaluate other linear transformer-based RS models and compared with HydraRec across various evaluation metrics. HydraRec outperforms other linear attention-based models as well as dot-product based attention models when used with causal masking for sequential recommendation next item prediction tasks. For bi-directional models its performance is comparable to the BERT4Rec model with an improvement in running time.

Fairness research in machine learning often centers on ensuring equitable performance of individual models. However, real-world recommendation systems are built on multiple models and even multiple stages, from candidate retrieval to scoring and serving, which raises challenges for responsible development and deployment. This system-level view, as highlighted by regulations like the EU AI Act, necessitates moving beyond auditing individual models as independent entities. We propose a holistic framework for modeling system-level fairness, focusing on the end-utility delivered to diverse user groups, and consider interactions between components such as retrieval and scoring models. We provide formal insights on the limitations of focusing solely on model-level fairness and highlight the need for alternative tools that account for heterogeneity in user preferences. To mitigate system-level disparities, we adapt closed-box optimization tools (e.g., BayesOpt) to jointly optimize utility and equity. We empirically demonstrate the effectiveness of our proposed framework on synthetic and real datasets, underscoring the need for a system-level framework.

How to Do It is Slate's sex advice column. Send it to Jessica and Rich here. I am a newly divorced bisexual dad who's moved to a city adjacent to my 20-year-old son's college. He's extremely shy and hasn't talked about sex with me in years. He identifies as queer but has provided no more detail than that.

Augmented Reality assistance are increasingly popular for supporting users with tasks like assembly and cooking. However, current practice typically provide reactive responses initialized from user requests, lacking consideration of rich contextual and user-specific information. To address this limitation, we propose a novel AR assistance system, Satori, that models both user states and environmental contexts to deliver proactive guidance. Our system combines the Belief-Desire-Intention (BDI) model with a state-of-the-art multi-modal large language model (LLM) to infer contextually appropriate guidance. The design is informed by two formative studies involving twelve experts. A sixteen within-subject study find that Satori achieves performance comparable to an designer-created Wizard-of-Oz (WoZ) system without relying on manual configurations or heuristics, thereby enhancing generalizability, reusability and opening up new possibilities for AR assistance.

The contextual multi-armed bandit (MAB) problem is crucial in sequential decision-making. A line of research, known as online clustering of bandits, extends contextual MAB by grouping similar users into clusters, utilizing shared features to improve learning efficiency. However, existing algorithms, which rely on the upper confidence bound (UCB) strategy, struggle to gather adequate statistical information to accurately identify unknown user clusters. As a result, their theoretical analyses require several strong assumptions about the "diversity" of contexts generated by the environment, leading to impractical settings, complicated analyses, and poor practical performance. Removing these assumptions has been a long-standing open problem in the clustering of bandits literature. In this paper, we provide two solutions to this open problem. First, following the i.i.d. context generation setting in existing studies, we propose two novel algorithms, UniCLUB and PhaseUniCLUB, which incorporate enhanced exploration mechanisms to accelerate cluster identification. Remarkably, our algorithms require substantially weaker assumptions while achieving regret bounds comparable to prior work. Second, inspired by the smoothed analysis framework, we propose a more practical setting that eliminates the requirement for i.i.d. context generation used in previous studies, thus enhancing the performance of existing algorithms for online clustering of bandits. Our technique can be applied to both graph-based and set-based clustering of bandits frameworks. Extensive evaluations on both synthetic and real-world datasets demonstrate that our proposed algorithms consistently outperform existing approaches.

The pace of smart home innovation hasn't slowed a whit in 2024, with new products such as the Amazon Echo Show 21 being unveiled just this week. The better news is that the smart home is no longer a niche market appealing only to enthusiasts willing to tolerate steep learning curves. We are, however, still living in a world of smart home silos; namely, Amazon Alexa, Apple HomeKit, and Google Home. The increasingly important Matter standard will unify those ecosystems one day, but that won't happen until Matter incorporates every smart home category. Considering that, we've picked more than one product in several categories, based primarily on which silo your smart home is in.

We experimented with various transformer-based language models, such as BERT [9], RoBERTa [10], and T5 [11] for these datasets to identify the most effective model based on our evaluation. After splitting the restructured datasets into a 90-10 ratio for training and testing, we proceeded to train the selected models (block highlighted in yellow) using a grid search of hyperparameters optimization. Given the multitude of possible tactics available at each proof state, we chose to provide multiple recommendations for the next proof step. To assess the accuracy of these recommendations (block highlighted in green), we use the n-correctness rate, which measures the likelihood that a correct tactic from the testing dataset is among the top-n recommended tactics, where n signifies the number of recommended tactics evaluated against the correct tactic. We found out that RoBERTa demonstrated superior performance across most cases for n = 7.

Let a digital sidekick take the strain. While user fatigue may be setting in – reports suggest a notable decline in usage – the world's biggest online dating company is launching an artificial intelligence assistant that it claims will "transform" online dating. Match Group, the technology company with the world's largest portfolio of dating platforms, has announced it is increasing investment in AI with new products coming in March 2025. An as yet unnamed AI assistant will perform core dating tasks such as selecting the photos it calculates will garner the most responses and recommend what prompts and information to put in a bio. It will also help a user choose the perfect partner.