Customer Lifetime Value (CLTV) prediction is a critical task in business applications. Accurately predicting CLTV is challenging in real-world business scenarios, as the distribution of CLTV is complex and mutable. Firstly, there is a large number of users without any consumption consisting of a long-tailed part that is too complex to fit. Secondly, the small set of high-value users spent orders of magnitude more than a typical user leading to a wide range of the CLTV distribution which is hard to capture in a single distribution. Existing approaches for CLTV estimation either assume a prior probability distribution and fit a single group of distribution-related parameters for all samples, or directly learn from the posterior distribution with manually predefined buckets in a heuristic manner. However, all these methods fail to handle complex and mutable distributions. In this paper, we propose a novel optimal distribution selection model OptDist for CLTV prediction, which utilizes an adaptive optimal sub-distribution selection mechanism to improve the accuracy of complex distribution modeling. Specifically, OptDist trains several candidate sub-distribution networks in the distribution learning module (DLM) for modeling the probability distribution of CLTV. Then, a distribution selection module (DSM) is proposed to select the sub-distribution for each sample, thus making the selection automatically and adaptively. Besides, we design an alignment mechanism that connects both modules, which effectively guides the optimization. We conduct extensive experiments on both two public and one private dataset to verify that OptDist outperforms state-of-the-art baselines. Furthermore, OptDist has been deployed on a large-scale financial platform for customer acquisition marketing campaigns and the online experiments also demonstrate the effectiveness of OptDist.

Multilingual large language models (MLLMs) struggle to answer questions posed in non-dominant languages, even though they have acquired the relevant knowledge from their dominant language corpus. In contrast, human multilinguals can overcome such non-native language context limitations through Positive Native Language Transfer (PNLT). Inspired by the process of PNLT, we analogize the dominant language of MLLMs to the native language of human multilinguals, and propose Native Language Prompting (NatLan) to simulate the PNLT observed in human multilinguals. It explicitly creates native language contexts for MLLMs to facilitate the elicitation of the rich native language knowledge during question-answering, unlocking the limitations imposed by non-native language contexts. By employing multi-MLLM collaboration, NatLan reduces the workload on each MLLM in simulating PNLT and refines semantic transfer. On the C-Eval benchmark, NatLan provides up to a 10.1% average accuracy improvement and up to a 5.0% increase in the hard-level subset across five MLLMs, surpassing all top-notch related methods. Our code is available at https://github.com/AnonyNLP/NatLan.

Data representation is a fundamental aspect of machine learning that significantly influences model performance, efficiency, and interpretability Rumelhart et al. (1986); Bengio et al. (2012); LeCun et al. (2015). In the era of deep learning, neural networks have become the primary tools for data representation in computer vision and natural language processing, leveraging their capacity to automatically extract features. For instance, neural networks trained on labeled data can serve as effective feature extractors when the final layer is removed Goodfellow et al. (2016). The core idea of transfer learning is to leverage learned representations from large upstream datasets to enhance the performance of target-specific downstream tasks. A particularly effective paradigm within transfer learning is pretraining followed by fine-tuning, which has gained increasing attention for its demonstrated efficiency in various studies Schroff et al. (2015); Dhillon et al. (2020); Chen et al. (2019, 2020c). During the pretraining phase, a representation is learned using a large, general dataset with annotations, which is then transferred to the target-specific task. In the fine-tuning stage, a relatively simple model is typically trained on the learned representation to address the specific problem at hand. There is a wide variety of transfer learning methods, along with corresponding theoretical guarantees, that have been proposed.

In the field of machine learning, comprehending the intricate training dynamics of neural networks poses a significant challenge. This paper explores the training dynamics of neural networks, particularly whether these dynamics can be expressed in a general closed-form solution. We demonstrate that the dynamics of the gradient flow in two-layer narrow networks is not an integrable system. Integrable systems are characterized by trajectories confined to submanifolds defined by level sets of first integrals (invariants), facilitating predictable and reducible dynamics. In contrast, non-integrable systems exhibit complex behaviors that are difficult to predict. To establish the non-integrability, we employ differential Galois theory, which focuses on the solvability of linear differential equations. We demonstrate that under mild conditions, the identity component of the differential Galois group of the variational equations of the gradient flow is non-solvable. This result confirms the system's non-integrability and implies that the training dynamics cannot be represented by Liouvillian functions, precluding a closed-form solution for describing these dynamics. Our findings highlight the necessity of employing numerical methods to tackle optimization problems within neural networks. The results contribute to a deeper understanding of neural network training dynamics and their implications for machine learning optimization strategies.

Statistical agencies rely on sampling techniques to collect socio-demographic data crucial for policy-making and resource allocation. This paper shows that surveys of important societal relevance introduce sampling errors that unevenly impact group-level estimates, thereby compromising fairness in downstream decisions. To address these issues, this paper introduces an optimization approach modeled on real-world survey design processes, ensuring sampling costs are optimized while maintaining error margins within prescribed tolerances. Additionally, privacy-preserving methods used to determine sampling rates can further impact these fairness issues. The paper explores the impact of differential privacy on the statistics informing the sampling process, revealing a surprising effect: not only the expected negative effect from the addition of noise for differential privacy is negligible, but also this privacy noise can in fact reduce unfairness as it positively biases smaller counts. These findings are validated over an extensive analysis using datasets commonly applied in census statistics.

To achieve dexterity comparable to that of humans, robots must intelligently process tactile sensor data. Taxel-based tactile signals often have low spatial-resolution, with non-standardized representations. In this paper, we propose a novel framework, HyperTaxel, for learning a geometrically-informed representation of taxel-based tactile signals to address challenges associated with their spatial resolution. We use this representation and a contrastive learning objective to encode and map sparse low-resolution taxel signals to high-resolution contact surfaces. To address the uncertainty inherent in these signals, we leverage joint probability distributions across multiple simultaneous contacts to improve taxel hyper-resolution. We evaluate our representation by comparing it with two baselines and present results that suggest our representation outperforms the baselines. Furthermore, we present qualitative results that demonstrate the learned representation captures the geometric features of the contact surface, such as flatness, curvature, and edges, and generalizes across different objects and sensor configurations. Moreover, we present results that suggest our representation improves the performance of various downstream tasks, such as surface classification, 6D in-hand pose estimation, and sim-to-real transfer.

Recent advances in AI have been significantly driven by the capabilities of large language models (LLMs) to solve complex problems in ways that resemble human thinking. However, there is an ongoing debate about the extent to which LLMs are capable of actual reasoning. Central to this debate are two key probabilistic concepts that are essential for connecting causes to their effects: the probability of necessity (PN) and the probability of sufficiency (PS). This paper introduces a framework that is both theoretical and practical, aimed at assessing how effectively LLMs are able to replicate real-world reasoning mechanisms using these probabilistic measures. By viewing LLMs as abstract machines that process information through a natural language interface, we examine the conditions under which it is possible to compute suitable approximations of PN and PS. Our research marks an important step towards gaining a deeper understanding of when LLMs are capable of reasoning, as illustrated by a series of math examples.

This paper introduces a reinforcement learning (RL) platform that enhances end-to-end user journeys in healthcare digital tools through personalization. We explore a case study with SwipeRx, the most popular all-in-one app for pharmacists in Southeast Asia, demonstrating how the platform can be used to personalize and adapt user experiences. Our RL framework is tested through a series of experiments with product recommendations tailored to each pharmacy based on real-time information on their purchasing history and in-app engagement, showing a significant increase in basket size. By integrating adaptive interventions into existing mobile health solutions and enriching user journeys, our platform offers a scalable solution to improve pharmaceutical supply chain management, health worker capacity building, and clinical decision and patient care, ultimately contributing to better healthcare outcomes.

In knowledge-intensive tasks such as open-domain question answering To overcome the limitations of LLMs' parametric knowledge, retrieval (OpenQA), Large Language Models (LLMs) often struggle augmented generation (RAG) [11, 27] is explored, equipping to generate factual answers relying solely on their internal (parametric) LLMs with a retriever to gather necessary evidence from external knowledge. To address this limitation, Retrieval-Augmented sources. Among the two components of RAG, improving the retriever Generation (RAG) systems enhance LLMs by retrieving relevant information is more feasible due to the recent trend of black-box APIs from external sources, thereby positioning the retriever [33] and the high cost and time requirements of fine-tuning opensource as a pivotal component. Although dense retrieval demonstrates LLMs [10]. The retriever, a critical part of RAG, is typically state-of-the-art performance, its training poses challenges due to either a traditional unsupervised retriever like BM25 [38] or a more the scarcity of ground-truth evidence, largely attributed to the high advanced neural retriever, such as dense retrieval [20, 21, 32, 51], costs of human annotation. In this paper, we propose W-RAG by which encodes questions and passages into the same embedding utilizing the ranking capabilities of LLMs to create weakly labeled space and then measures the question-passage relevance score by data for training dense retrievers.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.