Predicting Company Growth by Econophysics informed Machine Learning

Predicting company growth is crucial for strategic adjustment, operational decision-making, risk assessment, and loan eligibility reviews. Traditional models for company growth often focus too much on theory, overlooking practical forecasting, or they rely solely on time series forecasting techniques, ignoring interpretability and the inherent mechanisms of company growth. In this paper, we propose a machine learning-based prediction framework that incorporates an econophysics model for company growth. Our model captures both the intrinsic growth mechanisms of companies led by scaling laws and the fluctuations influenced by random factors and individual decisions, demonstrating superior predictive performance compared with methods that use time series techniques alone. Its advantages are more pronounced in long-range prediction tasks. By explicitly modeling the baseline growth and volatility components, our model is more interpretable.

Efficient Toxic Content Detection by Bootstrapping and Distilling Large Language Models

Toxic content detection is crucial for online services to remove inappropriate content that violates community standards. To automate the detection process, prior works have proposed varieties of machine learning (ML) approaches to train Language Models (LMs) for toxic content detection. However, both their accuracy and transferability across datasets are limited. Recently, Large Language Models (LLMs) have shown promise in toxic content detection due to their superior zero-shot and few-shot in-context learning ability as well as broad transferability on ML tasks. However, efficiently designing prompts for LLMs remains challenging. Moreover, the high run-time cost of LLMs may hinder their deployments in production. To address these challenges, in this work, we propose BD-LLM, a novel and efficient approach to Bootstrapping and Distilling LLMs for toxic content detection. Specifically, we design a novel prompting method named Decision-Tree-of-Thought (DToT) to bootstrap LLMs' detection performance and extract high-quality rationales. DToT can automatically select more fine-grained context to re-prompt LLMs when their responses lack confidence. Additionally, we use the rationales extracted via DToT to fine-tune student LMs. Our experimental results on various datasets demonstrate that DToT can improve the accuracy of LLMs by up to 4.6%. Furthermore, student LMs fine-tuned with rationales extracted via DToT outperform baselines on all datasets with up to 16.9\% accuracy improvement, while being more than 60x smaller than conventional LLMs. Finally, we observe that student LMs fine-tuned with rationales exhibit better cross-dataset transferability.

Neural Network Pruning by Gradient Descent

The rapid increase in the parameters of deep learning models has led to significant costs, challenging computational efficiency and model interpretability. In this paper, we introduce a novel and straightforward neural network pruning framework that incorporates the Gumbel-Softmax technique. This framework enables the simultaneous optimization of a network's weights and topology in an end-to-end process using stochastic gradient descent. Empirical results demonstrate its exceptional compression capability, maintaining high accuracy on the MNIST dataset with only 0.15\% of the original network parameters. Moreover, our framework enhances neural network interpretability, not only by allowing easy extraction of feature importance directly from the pruned network but also by enabling visualization of feature symmetry and the pathways of information propagation from features to outcomes. Although the pruning strategy is learned through deep learning, it is surprisingly intuitive and understandable, focusing on selecting key representative features and exploiting data patterns to achieve extreme sparse pruning. We believe our method opens a promising new avenue for deep learning pruning and the creation of interpretable machine learning systems.

Data driven modeling of self-similar dynamics

Multiscale modeling of complex systems is crucial for understanding their intricacies. Data-driven multiscale modeling has emerged as a promising approach to tackle challenges associated with complex systems. On the other hand, self-similarity is prevalent in complex systems, hinting that large-scale complex systems can be modeled at a reduced cost. In this paper, we introduce a multiscale neural network framework that incorporates self-similarity as prior knowledge, facilitating the modeling of self-similar dynamical systems. For deterministic dynamics, our framework can discern whether the dynamics are self-similar. For uncertain dynamics, it can compare and determine which parameter set is closer to self-similarity. The framework allows us to extract scale-invariant kernels from the dynamics for modeling at any scale. Moreover, our method can identify the power law exponents in self-similar systems. Preliminary tests on the Ising model yielded critical exponents consistent with theoretical expectations, providing valuable insights for addressing critical phase transitions in non-equilibrium systems.

Finding emergence in data: causal emergence inspired dynamics learning

Modelling complex dynamical systems in a data-driven manner is challenging due to the presence of emergent behaviors and properties that cannot be directly captured by micro-level observational data. Therefore, it is crucial to develop a model that can effectively capture emergent dynamics at the macro-level and quantify emergence based on the available data. Drawing inspiration from the theory of causal emergence, this paper introduces a machine learning framework aimed at learning macro-dynamics within an emergent latent space. The framework achieves this by maximizing the effective information (EI) to obtain a macro-dynamics model with stronger causal effects. Experimental results on both simulated and real data demonstrate the effectiveness of the proposed framework. Not only does it successfully capture emergent patterns, but it also learns the coarse-graining strategy and quantifies the degree of causal emergence in the data. Furthermore, experiments conducted on environments different from the training dataset highlight the superior generalization ability of our model.

Dynamic Graph Neural Network with Adaptive Edge Attributes for Air Quality Predictions

Air quality prediction is a typical spatio-temporal modeling problem, which always uses different components to handle spatial and temporal dependencies in complex systems separately. Previous models based on time series analysis and Recurrent Neural Network (RNN) methods have only modeled time series while ignoring spatial information. Previous GCNs-based methods usually require providing spatial correlation graph structure of observation sites in advance. The correlations among these sites and their strengths are usually calculated using prior information. However, due to the limitations of human cognition, limited prior information cannot reflect the real station-related structure or bring more effective information for accurate prediction. To this end, we propose a novel Dynamic Graph Neural Network with Adaptive Edge Attributes (DGN-AEA) on the message passing network, which generates the adaptive bidirected dynamic graph by learning the edge attributes as model parameters. Unlike prior information to establish edges, our method can obtain adaptive edge information through end-to-end training without any prior information. Thus reduced the complexity of the problem. Besides, the hidden structural information between the stations can be obtained as model by-products, which can help make some subsequent decision-making analyses. Experimental results show that our model received state-of-the-art performance than other baselines.

Extracting lung function-correlated information from CT-encoded static textures

The inherent characteristics of lung tissues, which are independent of breathing manoeuvre, may provide fundamental information on lung function. This paper attempted to study function-correlated lung textures and their spatial distribution from CT. 21 lung cancer patients with thoracic 4DCT scans, DTPA-SPECT ventilation images (V), and available pulmonary function test (PFT) measurements were collected. 79 radiomic features were included for analysis, and a sparse-to-fine strategy including subregional feature discovery and voxel-wise feature distribution study was carried out to identify the function-correlated radiomic features. At the subregion level, lung CT images were partitioned and labeled as defected/non-defected patches according to reference V. At the voxel-wise level, feature maps (FMs) of selected feature candidates were generated for each 4DCT phase. Quantitative metrics, including Spearman coefficient of correlation (SCC) and Dice similarity coefficient (DSC) for FM-V spatial agreement assessments, intra-class coefficient of correlation (ICC) for FM robustness evaluations, and FM-PFT comparisons, were applied to validate the results. At the subregion level, eight function-correlated features were filtered out with medium-to-large statistical strength (effect size>0.330) to differentiate defected/non-defected lung regions. At the voxel-wise level, FMs of candidates yielded moderate-to-strong voxel-wise correlations with reference V. Among them, FMs of GLDM Dependence Non-uniformity showed the highest robust (ICC=0.96) spatial correlation, with median SCCs ranging from 0.54 to 0.59 throughout ten phases. Its phase-averaged FM achieved a median SCC of 0.60, the median DSC of 0.60/0.65 for high/low functional lung volumes, respectively, and the correlation of 0.646 between the spatially averaged feature values and PFT measurements.

Multi-view Contrastive Learning with Additive Margin for Adaptive Nasopharyngeal Carcinoma Radiotherapy Prediction

The prediction of adaptive radiation therapy (ART) prior to radiation therapy (RT) for nasopharyngeal carcinoma (NPC) patients is important to reduce toxicity and prolong the survival of patients. Currently, due to the complex tumor micro-environment, a single type of high-resolution image can provide only limited information. Meanwhile, the traditional softmax-based loss is insufficient for quantifying the discriminative power of a model. To overcome these challenges, we propose a supervised multi-view contrastive learning method with an additive margin (MMCon). For each patient, four medical images are considered to form multi-view positive pairs, which can provide additional information and enhance the representation of medical images. In addition, the embedding space is learned by means of contrastive learning. NPC samples from the same patient or with similar labels will remain close in the embedding space, while NPC samples with different labels will be far apart. To improve the discriminative ability of the loss function, we incorporate a margin into the contrastive learning. Experimental result show this new learning objective can be used to find an embedding space that exhibits superior discrimination ability for NPC images.

How Much Privacy Does Federated Learning with Secure Aggregation Guarantee?

Federated learning (FL) has attracted growing interest for enabling privacy-preserving machine learning on data stored at multiple users while avoiding moving the data off-device. However, while data never leaves users' devices, privacy still cannot be guaranteed since significant computations on users' training data are shared in the form of trained local models. These local models have recently been shown to pose a substantial privacy threat through different privacy attacks such as model inversion attacks. As a remedy, Secure Aggregation (SA) has been developed as a framework to preserve privacy in FL, by guaranteeing the server can only learn the global aggregated model update but not the individual model updates. While SA ensures no additional information is leaked about the individual model update beyond the aggregated model update, there are no formal guarantees on how much privacy FL with SA can actually offer; as information about the individual dataset can still potentially leak through the aggregated model computed at the server. In this work, we perform a first analysis of the formal privacy guarantees for FL with SA. Specifically, we use Mutual Information (MI) as a quantification metric and derive upper bounds on how much information about each user's dataset can leak through the aggregated model update. When using the FedSGD aggregation algorithm, our theoretical bounds show that the amount of privacy leakage reduces linearly with the number of users participating in FL with SA. To validate our theoretical bounds, we use an MI Neural Estimator to empirically evaluate the privacy leakage under different FL setups on both the MNIST and CIFAR10 datasets. Our experiments verify our theoretical bounds for FedSGD, which show a reduction in privacy leakage as the number of users and local batch size grow, and an increase in privacy leakage with the number of training rounds.

Graph Auto-Encoders for Network Completion

Completing a graph means inferring the missing nodes and edges from a partially observed network. Different methods have been proposed to solve this problem, but none of them employed the pattern similarity of parts of the graph. In this paper, we propose a model to use the learned pattern of connections from the observed part of the network based on the Graph Auto-Encoder technique and generalize these patterns to complete the whole graph. Our proposed model achieved competitive performance with less information needed. Empirical analysis of synthetic datasets and real-world datasets from different domains show that our model can complete the network with higher accuracy compared with baseline prediction models in most cases. Furthermore, we also studied the character of the model and found it is particularly suitable to complete a network that has more complex local connection patterns.