Search

Time series forecasting aids decision-making, especially for stakeholders who rely on accurate predictions, making it very important to understand and explain these models to ensure informed decisions. Traditional explainable AI (XAI) methods, which underline feature or temporal importance, often require expert knowledge. In contrast, natural language explanations (NLEs) are more accessible to laypeople. However, evaluating forecast NLEs is difficult due to the complex causal relationships in time series data. To address this, we introduce two new performance metrics based on simulatability, assessing how well a human surrogate can predict model forecasts using the explanations. Experiments show these metrics differentiate good from poor explanations and align with human judgments. Utilizing these metrics, we further evaluate the ability of state-of-the-art large language models (LLMs) to generate explanations for time series data, finding that numerical reasoning, rather than model size, is the main factor influencing explanation quality.

Time series foundation models have demonstrated impressive performance as zero-shot forecasters. However, achieving effectively unified training on time series remains an open challenge. Existing approaches introduce some level of model specialization to account for the highly heterogeneous nature of time series data. For instance, Moirai pursues unified training by employing multiple input/output projection layers, each tailored to handle time series at a specific frequency. Similarly, TimesFM maintains a frequency embedding dictionary for this purpose. We identify two major drawbacks to this human-imposed frequency-level model specialization: (1) Frequency is not a reliable indicator of the underlying patterns in time series. For example, time series with different frequencies can display similar patterns, while those with the same frequency may exhibit varied patterns. (2) Non-stationarity is an inherent property of real-world time series, leading to varied distributions even within a short context window of a single time series. Frequency-level specialization is too coarse-grained to capture this level of diversity. To address these limitations, this paper introduces Moirai-MoE, using a single input/output projection layer while delegating the modeling of diverse time series patterns to the sparse mixture of experts (MoE) within Transformers. With these designs, Moirai-MoE reduces reliance on human-defined heuristics and enables automatic token-level specialization. Extensive experiments on 39 datasets demonstrate the superiority of Moirai-MoE over existing foundation models in both in-distribution and zero-shot scenarios. Furthermore, this study conducts comprehensive model analyses to explore the inner workings of time series MoE foundation models and provides valuable insights for future research.

Time series foundation models excel in zero-shot forecasting, handling diverse tasks without explicit training. However, the advancement of these models has been hindered by the lack of comprehensive benchmarks. To address this gap, we introduce the General Time Series Forecasting Model Evaluation, GIFT-Eval, a pioneering benchmark aimed at promoting evaluation across diverse datasets. GIFT-Eval encompasses 23 datasets over 144,000 time series and 177 million data points, spanning seven domains, 10 frequencies, multivariate inputs, and prediction lengths ranging from short to long-term forecasts. To facilitate the effective pretraining and evaluation of foundation models, we also provide a non-leaking pretraining dataset containing approximately 230 billion data points. Additionally, we provide a comprehensive analysis of 17 baselines, which includes statistical models, deep learning models, and foundation models. We discuss each model in the context of various benchmark characteristics and offer a qualitative analysis that spans both deep learning and foundation models. We believe the insights from this analysis, along with access to this new standard zero-shot time series forecasting benchmark, will guide future developments in time series foundation models. Code, data, and the leaderboard can be found at https://github.com/SalesforceAIResearch/gift-eval .

Click-through-rate (CTR) prediction plays an important role in online advertising and ad recommender systems. In the past decade, maximizing CTR has been the main focus of model development and solution creation. Therefore, researchers and practitioners have proposed various models and solutions to enhance the effectiveness of CTR prediction. Most of the existing literature focuses on capturing either implicit or explicit feature interactions. Although implicit interactions are successfully captured in some studies, explicit interactions present a challenge for achieving high CTR by extracting both low-order and high-order feature interactions. Unnecessary and irrelevant features may cause high computational time and low prediction performance. Furthermore, certain features may perform well with specific predictive models while underperforming with others. Also, feature distribution may fluctuate due to traffic variations. Most importantly, in live production environments, resources are limited, and the time for inference is just as crucial as training time. Because of all these reasons, feature selection is one of the most important factors in enhancing CTR prediction model performance. Simple filter-based feature selection algorithms do not perform well and they are not sufficient. An effective and efficient feature selection algorithm is needed to consistently filter the most useful features during live CTR prediction process. In this paper, we propose a heuristic algorithm named Neighborhood Search with Heuristic-based Feature Selection (NeSHFS) to enhance CTR prediction performance while reducing dimensionality and training time costs. We conduct comprehensive experiments on three public datasets to validate the efficiency and effectiveness of our proposed solution.

Deep learning has revolutionized biomedical research by providing sophisticated methods to handle complex, high-dimensional data. Multimodal deep learning (MDL) further enhances this capability by integrating diverse data types such as imaging, textual data, and genetic information, leading to more robust and accurate predictive models. In MDL, differently from early and late fusion methods, intermediate fusion stands out for its ability to effectively combine modality-specific features during the learning process. This systematic review aims to comprehensively analyze and formalize current intermediate fusion methods in biomedical applications. We investigate the techniques employed, the challenges faced, and potential future directions for advancing intermediate fusion methods. Additionally, we introduce a structured notation to enhance the understanding and application of these methods beyond the biomedical domain. Our findings are intended to support researchers, healthcare professionals, and the broader deep learning community in developing more sophisticated and insightful multimodal models. Through this review, we aim to provide a foundational framework for future research and practical applications in the dynamic field of MDL.

Current metrics for evaluating Dialogue State Tracking (DST) systems exhibit three primary limitations. They: i) erroneously presume a uniform distribution of slots throughout the dialog, ii) neglect to assign partial scores for individual turns, iii) frequently overestimate or underestimate performance by repeatedly counting the models' successful or failed predictions. To address these shortcomings, we introduce a novel metric: Granular Change Accuracy (GCA). GCA focuses on evaluating the predicted changes in dialogue state over the entire dialogue history. Benchmarking reveals that GCA effectively reduces biases arising from distribution uniformity and the positioning of errors across turns, resulting in a more precise evaluation. Notably, we find that these biases are particularly pronounced when evaluating few-shot or zero-shot trained models, becoming even more evident as the model's error rate increases. Hence, GCA offers significant promise, particularly for assessing models trained with limited resources. Our GCA implementation is a useful addition to the pool of DST metrics., Comment: Accepted to COLING 2024

We investigate the structural properties of binary linear codes whose permutation automorphism group has a fixed point free automorphism of order $3$. We prove that up to dimension or codimension $4$, there is no binary linear code whose permutation automorphism group is generated by a fixed point free permutation of order $3$. We also prove that there is no binary $5$-dimensional linear code whose length is at least $30$ and whose permutation automorphism group is generated by a fixed point free permutation of order $3$., Comment: 10 pages

Image coding for machines (ICM) aims at reducing the bitrate required to represent an image while minimizing the drop in machine vision analysis accuracy. In many use cases, such as surveillance, it is also important that the visual quality is not drastically deteriorated by the compression process. Recent works on using neural network (NN) based ICM codecs have shown significant coding gains against traditional methods; however, the decompressed images, especially at low bitrates, often contain checkerboard artifacts. We propose an effective decoder finetuning scheme based on adversarial training to significantly enhance the visual quality of ICM codecs, while preserving the machine analysis accuracy, without adding extra bitcost or parameters at the inference phase. The results show complete removal of the checkerboard artifacts at the negligible cost of -1.6% relative change in task performance score. In the cases where some amount of artifacts is tolerable, such as when machine consumption is the primary target, this technique can enhance both pixel-fidelity and feature-fidelity scores without losing task performance.

The Privacy Sandbox Attribution Reporting API has been recently deployed by Google Chrome to support the basic advertising functionality of attribution reporting (aka conversion measurement) after deprecation of third-party cookies. The API implements a collection of privacy-enhancing guardrails including contribution bounding and noise injection. It also offers flexibility for the analyst to allocate the contribution budget. In this work, we present methods for optimizing the allocation of the contribution budget for summary reports from the Attribution Reporting API. We evaluate them on real-world datasets as well as on a synthetic data model that we find to accurately capture real-world conversion data. Our results demonstrate that optimizing the parameters that can be set by the analyst can significantly improve the utility achieved by querying the API while satisfying the same privacy bounds.