Scaling laws have played a major role in the modern AI revolution, providing practitioners predictive power over how the model performance will improve with increasing data, compute, and number of model parameters. This has spurred an intense interest in the origin of neural scaling laws, with a common suggestion being that they arise from power law structure already present in the data. In this paper we study scaling laws for transformers trained to predict random walks (bigrams) on graphs with tunable complexity. We demonstrate that this simplified setting already gives rise to neural scaling laws even in the absence of power law structure in the data correlations. We further consider dialing down the complexity of natural language systematically, by training on sequences sampled from increasingly simplified generative language models, from 4,2,1-layer transformer language models down to language bigrams, revealing a monotonic evolution of the scaling exponents. Our results also include scaling laws obtained from training on random walks on random graphs drawn from Erdös-Renyi and scale-free Barabási-Albert ensembles. Finally, we revisit conventional scaling laws for language modeling, demonstrating that several essential results can be reproduced using 2 layer transformers with context length of 50, provide a critical analysis of various fits used in prior literature, demonstrate an alternative method for obtaining compute optimal curves as compared with current practice in published literature, and provide preliminary evidence that maximal update parameterization may be more parameter efficient than standard parameterization.

Here we elaborate on the details of using SNFs as a variational approximation of the posterior distribution of a variational autoencoder (V AE) [21] as presented in our last results section. All experiments were run using PyTorch 1.2 and on GTX1080Ti cards. NSF block consists of two subsequent NSF layers with intermediate swap layers. "Biased data" is defined by running local Metropolis MC in each of the two wells. "Unbiased data" is produced by running Metropolis MC with a large proposal step (standard The other settings are the same as in Table 1.

A.3 Formulation of bound constrained dual problem Proposition 1 . For any non-negative p, q, we generate a feasible p ˆ, ˆ q as follows. In Section 5.3, we describe that it can be helpful to regularize We also mention here a minor difference in derivations for convenience of readers. As expected, this term also appears in these other formulations [ 25, 42 ]. All experiments run on a single P100 GPU. This adjustment was not necessary for CNN experiments.

Wearable accelerometers and gyroscopes encode fine-grained behavioural signatures that can be exploited to re-identify users, making privacy protection essential for healthcare applications. We introduce C-AAE, a compressive anonymizing autoencoder that marries an Anonymizing AutoEncoder (AAE) with Adaptive Differential Pulse-Code Modulation (ADPCM). The AAE first projects raw sensor windows into a latent space that retains activity-relevant features while suppressing identity cues. ADPCM then differentially encodes this latent stream, further masking residual identity information and shrinking the bitrate. Experiments on the MotionSense and PAMAP2 datasets show that C-AAE cuts user re-identification F1 scores by 10-15 percentage points relative to AAE alone, while keeping activity-recognition F1 within 5 percentage points of the unprotected baseline. ADPCM also reduces data volume by roughly 75 %, easing transmission and storage overheads. These results demonstrate that C-AAE offers a practical route to balancing privacy and utility in continuous, sensor-based activity recognition for healthcare.

Automatic citation generation for sentences in a document or report is paramount for intelligence analysts, cybersecurity, news agencies, and education personnel. In this research, we investigate whether large language models (LLMs) are capable of generating references based on two forms of sentence queries: (a) Direct Queries, LLMs are asked to provide author names of the given research article, and (b) Indirect Queries, LLMs are asked to provide the title of a mentioned article when given a sentence from a different article. To demonstrate where LLM stands in this task, we introduce a large dataset called REASONS comprising abstracts of the 12 most popular domains of scientific research on arXiv. From around 20K research articles, we make the following deductions on public and proprietary LLMs: (a) State-of-the-art, often called anthropomorphic GPT-4 and GPT-3.5, suffers from high pass percentage (PP) to minimize the hallucination rate (HR). When tested with Perplexity.ai (7B), they unexpectedly made more errors; (b) Augmenting relevant metadata lowered the PP and gave the lowest HR; (c) Advance retrieval-augmented generation (RAG) using Mistral demonstrates consistent and robust citation support on indirect queries and matched performance to GPT-3.5 and GPT-4. The HR across all domains and models decreased by an average of 41.93%, and the PP was reduced to 0% in most cases. In terms of generation quality, the average F1 Score and BLEU were 68.09% and 57.51%, respectively; (d) Testing with adversarial samples showed that LLMs, including the Advance RAG Mistral, struggle to understand context, but the extent of this issue was small in Mistral and GPT-4-Preview. Our study contributes valuable insights into the reliability of RAG for automated citation generation tasks.

Those can be competitive with deep neural networks on tabular data and, under some conditions, explainable. The explainability depends on the depth of the tree and the accuracy in each leaf of the tree. Decision trees containing leaves with unbalanced accuracy can provide misleading explanations. Low-accuracy leaves give less valid explanations, which could be interpreted as unfairness among explanations. Here, we train a shallow tree with the objective of minimizing the maximum misclassification error across each leaf node. Then, we extend each leaf with a separate tree-based model. The shallow tree provides a global explanation, while the overall statistical performance of the shallow tree with extended leaves improves upon decision trees of unlimited depth trained using classical methods (e.g., CART) and is comparable to state-of-the-art methods (e.g., well-tuned XGBoost).

To enhance documentation and maintenance practices, developers conventionally establish links between related software artifacts manually. Empirical research has revealed that developers frequently overlook this practice, resulting in significant information loss. To address this issue, automatic link recovery techniques have been proposed. However, these approaches primarily focused on improving prediction accuracy on randomly-split datasets, with limited attention given to the impact of data leakage and the generalizability of the predictive models. LinkFormer seeks to address these limitations. Our approach not only preserves and improves the accuracy of existing predictions but also enhances their alignment with real-world settings and their generalizability. First, to better utilize contextual information for prediction, we employ the Transformer architecture and fine-tune multiple pre-trained models on both textual and metadata information of issues and commits. Next, to gauge the effect of time on model performance, we employ two splitting policies during both the training and testing phases; randomly- and temporally-split datasets. Finally, in pursuit of a generic model that can demonstrate high performance across a range of projects, we undertake additional fine-tuning of LinkFormer within two distinct transfer-learning settings. Our findings support that to simulate real-world scenarios effectively, researchers must maintain the temporal flow of data when training models. Furthermore, the results demonstrate that LinkFormer outperforms existing methodologies by a significant margin, achieving a 48% improvement in F1-measure within a project-based setting. Finally, the performance of LinkFormer in the cross-project setting is comparable to its average performance within the project-based scenario.

Effective quantification of uncertainty is an essential and still missing step towards a greater adoption of deep-learning approaches in different applications, including mission-critical ones. In particular, investigations on the predictive uncertainty of deep-learning models describing non-linear dynamical systems are very limited to date. This paper is aimed at filling this gap and presents preliminary results on uncertainty quantification for system identification with neural state-space models. We frame the learning problem in a Bayesian probabilistic setting and obtain posterior distributions for the neural network's weights and outputs through approximate inference techniques. Based on the posterior, we construct credible intervals on the outputs and define a surprise index which can effectively diagnose usage of the model in a potentially dangerous out-of-distribution regime, where predictions cannot be trusted.