Current parameter-efficient fine-tuning methods for adapting pre-trained language models to downstream tasks are susceptible to interference from noisy data. Conventional noise-handling approaches either rely on laborious data pre-processing or employ model architecture modifications prone to error accumulation. In contrast to existing noise-process paradigms, we propose a noise-robust adaptation method via asymmetric LoRA poisoning experts (LoPE), a novel framework that enhances model robustness to noise only with generated noisy data. Drawing inspiration from the mixture-of-experts architecture, LoPE strategically integrates a dedicated poisoning expert in an asymmetric LoRA configuration. Through a two-stage paradigm, LoPE performs noise injection on the poisoning expert during fine-tuning to enhance its noise discrimination and processing ability. During inference, we selectively mask the dedicated poisoning expert to leverage purified knowledge acquired by normal experts for noise-robust output. Extensive experiments demonstrate that LoPE achieves strong performance and robustness purely through the low-cost noise injection, which completely eliminates the requirement of data cleaning.

Unsupervised classification is a fundamental machine learning problem. Real-world data often contain imperfections, characterized by uncertainty and imprecision, which are not well handled by traditional methods. Evidential clustering, based on Dempster-Shafer theory, addresses these challenges. This paper explores the underexplored problem of explaining evidential clustering results, which is crucial for high-stakes domains such as healthcare. Our analysis shows that, in the general case, representativity is a necessary and sufficient condition for decision trees to serve as abductive explainers. Building on the concept of representativity, we generalize this idea to accommodate partial labeling through utility functions. These functions enable the representation of "tolerable" mistakes, leading to the definition of evidential mistakeness as explanation cost and the construction of explainers tailored to evidential classifiers. Finally, we propose the Iterative Evidential Mistake Minimization (IEMM) algorithm, which provides interpretable and cautious decision tree explanations for evidential clustering functions. We validate the proposed algorithm on synthetic and real-world data. Taking into account the decision-maker's preferences, we were able to provide an explanation that was satisfactory up to 93% of the time.

In this paper, we investigate preference-based reinforcement learning (PbRL) that allows reinforcement learning (RL) agents to learn from human feedback. This is particularly valuable when defining a fine-grain reward function is not feasible. However, this approach is inefficient and impractical for promoting deep exploration in hard-exploration tasks with long horizons and sparse rewards. To tackle this issue, we introduce LOPE: Learning Online with trajectory Preference guidancE, an end-to-end preference-guided RL framework that enhances exploration efficiency in hard-exploration tasks. Our intuition is that LOPE directly adjusts the focus of online exploration by considering human feedback as guidance, avoiding learning a separate reward model from preferences. Specifically, LOPE includes a two-step sequential policy optimization process consisting of trust-region-based policy improvement and preference guidance steps. We reformulate preference guidance as a novel trajectory-wise state marginal matching problem that minimizes the maximum mean discrepancy distance between the preferred trajectories and the learned policy. Furthermore, we provide a theoretical analysis to characterize the performance improvement bound and evaluate the LOPE's effectiveness. When assessed in various challenging hard-exploration environments, LOPE outperforms several state-of-the-art methods regarding convergence rate and overall performance. The code used in this study is available at \url{https://github.com/buaawgj/LOPE}.

Short- and long-term outcomes of an algorithm often differ, with damaging downstream effects. A known example is a click-bait algorithm, which may increase short-term clicks but damage long-term user engagement. A possible solution to estimate the long-term outcome is to run an online experiment or A/B test for the potential algorithms, but it takes months or even longer to observe the long-term outcomes of interest, making the algorithm selection process unacceptably slow. This work thus studies the problem of feasibly yet accurately estimating the long-term outcome of an algorithm using only historical and short-term experiment data. Existing approaches to this problem either need a restrictive assumption about the short-term outcomes called surrogacy or cannot effectively use short-term outcomes, which is inefficient. Therefore, we propose a new framework called Long-term Off-Policy Evaluation (LOPE), which is based on reward function decomposition. LOPE works under a more relaxed assumption than surrogacy and effectively leverages short-term rewards to substantially reduce the variance. Synthetic experiments show that LOPE outperforms existing approaches particularly when surrogacy is severely violated and the long-term reward is noisy. In addition, real-world experiments on large-scale A/B test data collected on a music streaming platform show that LOPE can estimate the long-term outcome of actual algorithms more accurately than existing feasible methods.

Machine learning models have become increasingly popular for predicting the results of soccer matches, however, the lack of publicly-available benchmark datasets has made model evaluation challenging. The 2023 Soccer Prediction Challenge required the prediction of match results first in terms of the exact goals scored by each team, and second, in terms of the probabilities for a win, draw, and loss. The original training set of matches and features, which was provided for the competition, was augmented with additional matches that were played between 4 April and 13 April 2023, representing the period after which the training set ended, but prior to the first matches that were to be predicted (upon which the performance was evaluated). A CatBoost model was employed using pi-ratings as the features, which were initially identified as the optimal choice for calculating the win/draw/loss probabilities. Notably, deep learning models have frequently been disregarded in this particular task. Therefore, in this study, we aimed to assess the performance of a deep learning model and determine the optimal feature set for a gradient-boosted tree model. The model was trained using the most recent five years of data, and three training and validation sets were used in a hyperparameter grid search. The results from the validation sets show that our model had strong performance and stability compared to previously published models from the 2017 Soccer Prediction Challenge for win/draw/loss prediction.

VR technology has come a long way from the early Virtuality systems that inhabited our local malls in the '80s and '90s, with modern headsets offering 4K resolution, Dolby Atmos surround sound, and motion-sensing controllers. "If you want to feel these big sensations, you've got to have the infrastructure first," University of Chicago PhD student, Romain Nith, told Engadget. "You've got to go to theme parks, ride roller coasters, or you need bungie cords pulling you from the ceiling." And while the sensations are really like what they're simulating (because you're really being thrown around), "you can't have that in your living room." The JumpMod Haptic Backpack prototype, on the other hand, can effectively fool its user's sense of proprioception to make jumping in VR feel much more lifelike with a device the size of, well, a backpack.

This article is part of a VB Lab Insights series on AI sponsored by Microsoft and Nvidia. Don't miss additional articles in this series providing new industry insights, trends and analysis on how AI is transforming organizations. Scaling artificial intelligence (AI) is tough in any industry. And healthcare ranks among the toughest, thanks to highly complex applications, scattered stakeholder networks, stringent licensing and regulations, data privacy and security -- and the life-and-death nature of the industry. "If you mis-forecast an inventory level because your AI doesn't work, that's not great, but you'll recover," says Peter Durlach, Executive Vice President and Chief Strategy Officer of Nuance Communications, a conversational AI company specializing in healthcare. "If your clinical AI makes a mistake, like missing a cancerous nodule on an X-ray, that can have more serious consequences." Even with the current willingness of many organizations to fund AI initiatives, many healthcare organizations lack the skilled staff, technical know-how and bandwidth to deploy and scale AI into clinical workflows.

Digital generated image of bitcoin sign over glowing digital circuit board. Artificial intelligence (AI) crypto tokens are soaring in price this week, but price movements seem to be more of a crypto proxy to the AI bubble. The rally comes as a J.P Morgan report that says traders are turning their attention to AI and away from blockchain. "The rise in the price of AI-related cryptocurrencies can without a doubt be driven by real and tangible developments in the AI and blockchain industries," says Vasco Lopes, blockchain and artificial intelligence researcher at the NOVA school of technology near Lisbon, Portugal. "However, AI-related cryptocurrencies are also influenced by hype and investor sentiment, as the increased popularity of AI and AI-related products, such as the release of OpenAI's ChatGPT language model, generates excitement and interest in the AI sector." AI cryptos have reached a $4.27 billion market cap, up 56% from last week.

Lost or misattributed works by some of the finest writers of Spain's Golden Age could be discovered thanks to pioneering AI technology that has been used to identify a previously unknown play by the wildly prolific dramatist, poet, sailor and priest Lope de Vega. This week Spain's National Library announced that researchers trawling its massive archive had stumbled upon and verified a play that Lope is believed to have written a few years before his death in 1635. Like many plays of the Spanish Golden Age – the 16th- and 17th-century cultural boom that accompanied Spain's imperial growth and which birthed masterpieces by Lope, Cervantes, Calderón and Velázquez, among many others – La francesa Laura (The Frenchwoman Laura) is a tale of love, jealousy and social hierarchy in which suspicion demands an innocent woman be sacrificed on the altar of her husband's honour. But, unlike many similar plays of the period, Laura survives and the third act ends happily. Equally unusual was the manner of the play's discovery.

Neural Architecture Search (NAS) methods have been successfully applied to image tasks with excellent results. However, NAS methods are often complex and tend to converge to local minima as soon as generated architectures seem to yield good results. This paper proposes GEA, a novel approach for guided NAS. GEA guides the evolution by exploring the search space by generating and evaluating several architectures in each generation at initialisation stage using a zero-proxy estimator, where only the highest-scoring architecture is trained and kept for the next generation. Subsequently, GEA continuously extracts knowledge about the search space without increased complexity by generating several off-springs from an existing architecture at each generation. More, GEA forces exploitation of the most performant architectures by descendant generation while simultaneously driving exploration through parent mutation and favouring younger architectures to the detriment of older ones. Experimental results demonstrate the effectiveness of the proposed method, and extensive ablation studies evaluate the importance of different parameters. Results show that GEA achieves state-of-the-art results on all data sets of NAS-Bench-101, NAS-Bench-201 and TransNAS-Bench-101 benchmarks.