Human mobility behaviours are closely linked to various important societal problems such as traffic congestion, and epidemic control. However, collecting mobility data can be prohibitively expensive and involves serious privacy issues, posing a pressing need for high-quality generative mobility models. Previous efforts focus on learning the behaviour distribution from training samples, and generate new mobility data by sampling the learned distributions. They cannot effectively capture the coherent intentions that drive mobility behavior, leading to low sample efficiency and semantic-awareness. Inspired by the emergent reasoning ability in LLMs, we propose a radical perspective shift that reformulates mobility generation as a commonsense reasoning problem. In this paper, we design a novel Mobility Generation as Reasoning (MobiGeaR) framework that prompts LLM to recursively generate mobility behaviour. Specifically, we design a context-aware chain-of-thoughts prompting technique to align LLMs with context-aware mobility behaviour by few-shot in-context learning. Besides, MobiGeaR employ a divide-and-coordinate mechanism to exploit the synergistic effect between LLM reasoning and mechanistic gravity model. It leverages the step-by-step LLM reasoning to recursively generate a temporal template of activity intentions, which are then mapped to physical locations with a mechanistic gravity model. Experiments on two real-world datasets show MobiGeaR achieves state-of-the-art performance across all metrics, and substantially reduces the size of training samples at the same time. Besides, MobiGeaR also significantly improves the semantic-awareness of mobility generation by improving the intention accuracy by 62.23% and the generated mobility data is proven effective in boosting the performance of downstream applications. The implementation of our approach is available in the paper.

Cognitive Diagnosis (CD) algorithms receive growing research interest in intelligent education. Typically, these CD algorithms assist students by inferring their abilities (i.e., their proficiency levels on various knowledge concepts). The proficiency levels can enable further targeted skill training and personalized exercise recommendations, thereby promoting students' learning efficiency in online education. Recently, researchers have found that building and incorporating a student-exercise bipartite graph is beneficial for enhancing diagnostic performance. However, there are still limitations in their studies. On one hand, researchers overlook the heterogeneity within edges, where there can be both correct and incorrect answers. On the other hand, they disregard the uncertainty within edges, e.g., a correct answer can indicate true mastery or fortunate guessing. To address the limitations, we propose Adaptive Semantic-aware Graph-based Cognitive Diagnosis model (ASG-CD), which introduces a novel and effective way to leverage bipartite graph information in CD. Specifically, we first map students, exercises, and knowledge concepts into a latent representation space and combine these latent representations to obtain student abilities and exercise difficulties. After that, we propose a Semantic-aware Graph Neural Network Layer to address edge heterogeneity. This layer splits the original bipartite graph into two subgraphs according to edge semantics, and aggregates information based on these two subgraphs separately. To mitigate the impact of edge uncertainties, we propose an Adaptive Edge Differentiation Layer that dynamically differentiates edges, followed by keeping reliable edges and filtering out uncertain edges. Extensive experiments on three real-world datasets have demonstrated the effectiveness of ASG-CD.

The Continuous Learning (CL) paradigm consists of continuously evolving the parameters of the Deep Neural Network (DNN) model to progressively learn to perform new tasks without reducing the performance on previous tasks, i.e., avoiding the so-called catastrophic forgetting. However, the DNN parameter update in CL-based autonomous systems is extremely resource-hungry. The existing DNN accelerators cannot be directly employed in CL because they only support the execution of the forward propagation. Only a few prior architectures execute the backpropagation and weight update, but they lack the control and management for CL. Towards this, we design a hardware architecture, TinyCL, to perform CL on resource-constrained autonomous systems. It consists of a processing unit that executes both forward and backward propagation, and a control unit that manages memory-based CL workload. To minimize the memory accesses, the sliding window of the convolutional layer moves in a snake-like fashion. Moreover, the Multiply-and-Accumulate units can be reconfigured at runtime to execute different operations. As per our knowledge, our proposed TinyCL represents the first hardware accelerator that executes CL on autonomous systems. We synthesize the complete TinyCL architecture in a 65 nm CMOS technology node with the conventional ASIC design flow. It executes 1 epoch of training on a Conv + ReLU + Dense model on the CIFAR10 dataset in 1.76 s, while 1 training epoch of the same model using an Nvidia Tesla P100 GPU takes 103 s, thus achieving a 58 x speedup, consuming 86 mW in a 4.74 mm2 die.

Pretext training followed by task-specific fine-tuning has been a successful approach in vision and language domains. This paper proposes a self-supervised pretext training framework tailored to event sequence data. We introduce a novel alignment verification task that is specialized to event sequences, building on good practices in masked reconstruction and contrastive learning. Our pretext tasks unlock foundational representations that are generalizable across different down-stream tasks, including next-event prediction for temporal point process models, event sequence classification, and missing event interpolation. Experiments on popular public benchmarks demonstrate the potential of the proposed method across different tasks and data domains.

Large Language Models (LLMs) have demonstrated remarkable versatility in recent years, offering potential applications across specialized domains such as healthcare and medicine. Despite the availability of various open-source LLMs tailored for health contexts, adapting general-purpose LLMs to the medical domain presents significant challenges. In this paper, we introduce BioMistral, an open-source LLM tailored for the biomedical domain, utilizing Mistral as its foundation model and further pre-trained on PubMed Central. We conduct a comprehensive evaluation of BioMistral on a benchmark comprising 10 established medical question-answering (QA) tasks in English. We also explore lightweight models obtained through quantization and model merging approaches. Our results demonstrate BioMistral's superior performance compared to existing open-source medical models and its competitive edge against proprietary counterparts. Finally, to address the limited availability of data beyond English and to assess the multilingual generalization of medical LLMs, we automatically translated and evaluated this benchmark into 7 other languages. This marks the first large-scale multilingual evaluation of LLMs in the medical domain. Datasets, multilingual evaluation benchmarks, scripts, and all the models obtained during our experiments are freely released.

Oftentimes, machine learning applications using neural networks involve solving discrete optimization problems, such as in pruning, parameter-isolation-based continual learning and training of binary networks. Still, these discrete problems are combinatorial in nature and are also not amenable to gradient-based optimization. Additionally, classical approaches used in discrete settings do not scale well to large neural networks, forcing scientists and empiricists to rely on alternative methods. Among these, two main distinct sources of top-down information can be used to lead the model to good solutions: (1) extrapolating gradient information from points outside of the solution set (2) comparing evaluations between members of a subset of the valid solutions. We take continuation path (CP) methods to represent using purely the former and Monte Carlo (MC) methods to represent the latter, while also noting that some hybrid methods combine the two. The main goal of this work is to compare both approaches. For that purpose, we first overview the two classes while also discussing some of their drawbacks analytically. Then, on the experimental section, we compare their performance, starting with smaller microworld experiments, which allow more fine-grained control of problem variables, and gradually moving towards larger problems, including neural network regression and neural network pruning for image classification, where we additionally compare against magnitude-based pruning.

Recent work has shown the immense potential of synthetically generated datasets for training large language models (LLMs), especially for acquiring targeted skills. Current large-scale math instruction tuning datasets such as MetaMathQA (Yu et al., 2024) and MAmmoTH (Yue et al., 2024) are constructed using outputs from closed-source LLMs with commercially restrictive licenses. A key reason limiting the use of open-source LLMs in these data generation pipelines has been the wide gap between the mathematical skills of the best closed-source LLMs, such as GPT-4, and the best open-source LLMs. Building on the recent progress in open-source LLMs, our proposed prompting novelty, and some brute-force scaling, we construct OpenMathInstruct-1, a math instruction tuning dataset with 1.8M problem-solution pairs. The dataset is constructed by synthesizing code-interpreter solutions for GSM8K and MATH, two popular math reasoning benchmarks, using the recently released and permissively licensed Mixtral model. Our best model, OpenMath-CodeLlama-70B, trained on a subset of OpenMathInstruct-1, achieves a score of 84.6% on GSM8K and 50.7% on MATH, which is competitive with the best gpt-distilled models. We release our code, models, and the OpenMathInstruct-1 dataset under a commercially permissive license.

Random features models play a distinguished role in the theory of deep learning, describing the behavior of neural networks close to their infinite-width limit. In this work, we present a thorough analysis of the generalization performance of random features models for generic supervised learning problems with Gaussian data. Our approach, built with tools from the statistical mechanics of disordered systems, maps the random features model to an equivalent polynomial model, and allows us to plot average generalization curves as functions of the two main control parameters of the problem: the number of random features $N$ and the size $P$ of the training set, both assumed to scale as powers in the input dimension $D$. Our results extend the case of proportional scaling between $N$, $P$ and $D$. They are in accordance with rigorous bounds known for certain particular learning tasks and are in quantitative agreement with numerical experiments performed over many order of magnitudes of $N$ and $P$. We find good agreement also far from the asymptotic limits where $D\to \infty$ and at least one between $P/D^K$, $N/D^L$ remains finite.

Consider a predictor, a learner, whose input is a stream of discrete items. The predictor's task, at every time point, is probabilistic multiclass prediction, i.e., to predict which item may occur next by outputting zero or more candidate items, each with a probability, after which the actual item is revealed and the predictor learns from this observation. To output probabilities, the predictor keeps track of the proportions of the items it has seen. The predictor has constant (limited) space and we seek efficient prediction and update techniques: The stream is unbounded, the set of items is unknown to the predictor and their totality can also grow unbounded. Moreover, there is non-stationarity: the underlying frequencies of items may change, substantially, from time to time. For instance, new items may start appearing and a few currently frequent items may cease to occur again. The predictor, being space-bounded, need only provide probabilities for those items with (currently) sufficiently high frequency, i.e., the salient items. This problem is motivated in the setting of prediction games, a self-supervised learning regime where concepts serve as both the predictors and the predictands, and the set of concepts grows over time, resulting in non-stationarities as new concepts are generated and used. We develop moving average techniques designed to respond to such non-stationarities in a timely manner, and explore their properties. One is a simple technique based on queuing of count snapshots, and another is a combination of queuing together with an extended version of sparse EMA. The latter combination supports predictand-specific dynamic learning rates. We find that this flexibility allows for a more accurate and timely convergence.

Many recent works have studied the eigenvalue spectrum of the Conjugate Kernel (CK) defined by the nonlinear feature map of a feedforward neural network. However, existing results only establish weak convergence of the empirical eigenvalue distribution, and fall short of providing precise quantitative characterizations of the ''spike'' eigenvalues and eigenvectors that often capture the low-dimensional signal structure of the learning problem. In this work, we characterize these signal eigenvalues and eigenvectors for a nonlinear version of the spiked covariance model, including the CK as a special case. Using this general result, we give a quantitative description of how spiked eigenstructure in the input data propagates through the hidden layers of a neural network with random weights. As a second application, we study a simple regime of representation learning where the weight matrix develops a rank-one signal component over training and characterize the alignment of the target function with the spike eigenvector of the CK on test data.