This article leverages deep reinforcement learning (DRL) to hedge American put options, utilizing the deep deterministic policy gradient (DDPG) method. The agents are first trained and tested with Geometric Brownian Motion (GBM) asset paths and demonstrate superior performance over traditional strategies like the Black-Scholes (BS) Delta, particularly in the presence of transaction costs. To assess the real-world applicability of DRL hedging, a second round of experiments uses a market calibrated stochastic volatility model to train DRL agents. Specifically, 80 put options across 8 symbols are collected, stochastic volatility model coefficients are calibrated for each symbol, and a DRL agent is trained for each of the 80 options by simulating paths of the respective calibrated model. Not only do DRL agents outperform the BS Delta method when testing is conducted using the same calibrated stochastic volatility model data from training, but DRL agents achieves better results when hedging the true asset path that occurred between the option sale date and the maturity. As such, not only does this study present the first DRL agents tailored for American put option hedging, but results on both simulated and empirical market testing data also suggest the optimality of DRL agents over the BS Delta method in real-world scenarios. Finally, note that this study employs a model-agnostic Chebyshev interpolation method to provide DRL agents with option prices at each time step when a stochastic volatility model is used, thereby providing a general framework for an easy extension to more complex underlying asset processes.

The COVID-19 pandemic highlighted the importance of effective air disinfection technologies to mitigate the spread of airborne pathogens. In-duct ultraviolet germicidal irradiation (UVGI) systems may be a viable solution. System performance should be validated using biodosimetry, as per several existing standards. These tests yield the kill rates of a surrogate organism and its reduction equivalent dose (RED), with the intent that the RED be extrapolated to a predicted kill rate of a target pathogen of interest, such as SARS-CoV-2. However, this extrapolation requires adjustments to account for potential bias between the surrogate RED and the target RED (called the RED bias). Overlooking this mismatch can lead to inaccurate claims of the actual inactivation performance against the target. This study uses computational fluid dynamics modeling to analyze the UV dose distribution and resulting RED bias in in-duct UVGI systems. The results showed that, when MS2, a UV-resistant organism, is used as a surrogate to predict SARS-CoV-2 inactivation efficiency, the RED bias ranged from 1.14 to 1.46 within the studied cases, suggesting that the SARS-CoV-2 log inactivation can be overestimated by as much as 46%. This study also explores the combined variable (CV) approach as a more accurate method for predicting pathogen inactivation, offering an alternative to the RED bias approach. Both the RED bias approach and the CV approach were effective in improving the accuracy of performance predictions. This study underscores the need for the industry to incorporate considerations of the RED bias phenomenon in the future development of performance evaluation guidance to avoid overestimation of the treatment performance and safeguard public health.

In this article, we extend a statistical test of graph clusterability, the δ test, to directed graphs with no self loops. The δ test, originally designed for undirected graphs, is based on the premise that graphs with a clustered structure display a mean local density that is statistically higher than the graph's global density. We posit that graphs that do not meet this necessary (but not sufficient) condition for clusterability can be considered unsuited to clustering. In such cases, vertex clusters do not offer a meaningful summary of the broader graph. Additionally in this study, we aim to determine the optimal sample size (number of neighborhoods). Our test, designed for the analysis of large networks, is based on sampling subsets of neighborhoods/nodes. It is designed for cases where computing the density of every node's neighborhood is infeasible. Our results show that the δ test performs very well, even with very small samples of neighborhoods (1%). It accurately detects unclusterable graphs and is also shown to be robust to departures from the underlying assumptions of the t test.

Chlorine-based disinfectants are often used to achieve the required microbiological effluent limits and to ensure appropriate public health protection against waterborne pathogens; however, they can produce potentially harmful disinfection by-products and negatively affect the ecosystem of effluent receiving bodies. Peracetic Acid (PAA) and Performic Acid (PFA) are emerging as promising disinfectants in water treatment due to their effectiveness against a wide spectrum of microbes and minimal environmental impact, addressing some of the limitations associated with traditional chlorine-based disinfectants. In present work, we present and analyze the secondary effluent wastewater bench test results for demand, decay, and microbial inactivation kinetics of PAA, PFA, and sodium hypochlorite (SH) from six North American water resource recovery facilities (WRRFs) using an advanced fitting method based on uncertainty ellipses. To evaluate the effectiveness and suitability of alternative disinfectants, the inactivation performance of the PAA and PFA against E. coli, fecal coliform and enterococci were investigated and compared to the inactivation performance of SH. When comparing the integral-CT or ICT (i.e., integral of disinfectant concentration over time) based exposure performance of PFA against SH and PAA, it exhibits the superior inactivation performance in the limited ICT values for all the fecal indicator bacteria. For instance, to achieve a 2-log10 reduction of E. coli inactivation, ICT required for PFA ranges from approximately 1 to 3.6 mg min/L, for PAA from 8.2 to 21 mg min/L and for SH from 0.6 to 22 mg min/L across six WRRFs. Additionally, the residual concentration of the disinfectant dosage was examined to understand the uncertainties in demand/decay model development. Developing a confidence region provides guidance for conducting future experiments to improve the accuracy and reliability of subsequent tests. Later, the results of demand/decay and inactivation kinetics are utilized in process modeling, considering four dosing control schemes. Our findings indicate that the advanced dosing scheme, which accounts for wastewater variability and the hydraulic characteristics of the contact chamber, reduces disinfectant usage by up to 36% compared to the conventional flow pacing strategy while ensuring compliance with regulatory standards. Additionally, the advanced control scheme demonstrated steady and reliable performance in achieving target microbial limits throughout its operation.

This paper investigates the use of reinforcement learning (RL) algorithms to learn adaptive hedging strategies for derivatives under realistic market conditions, incorporating permanent market impact, execution slippage, and transaction costs. Market frictions arising from trading have been explored in the optimal trade execution literature; however, their influence on derivative hedging strategies remains comparatively understudied within RL contexts. Traditional hedging methods have typically assumed frictionless markets with only transaction costs. We illustrate that the dynamic decision problem posed by hedging with frictions can be modelled effectively with RL, demonstrating efficacy across various market frictions to minimize hedging losses. The results include a comparative analysis of the performance of three RL models across simulated price paths, demonstrating their varying effectiveness and adaptability in these friction-intensive environments. We find that RL agents, specifically TD3 and SAC, can outperform traditional delta hedging strategies in both simplistic and complex, illiquid environments highlighted by 2/3rd reductions in expected hedging losses and over 50% reductions in 5th percentile conditional value at risk (CVaR). These findings demonstrate that DRL agents can serve as a valuable risk management tool for financial institutions, especially given their adaptability to different market conditions and securities.

Chlorine-based disinfectants are widely used in North America and can achieve the desired microbiological limits. However, these disinfectants are highly toxic and produce potentially hazardous by-products. The present work examines the efficacy of PFA against a wide range of fecal indicators and compares its performance against peracetic acid and sodium hypochlorite. Later, the results of demand/decay and inactivation kinetics are employed in the process modelling where we have considered four dosing control schemes. The result anticipates that the advanced dosing scheme, accounting for wastewater variability and hydraulic characteristics in the contact chamber, demands a lesser disinfectant while meeting regulatory compliance. In addition, the advance control scheme demonstrated steady and reliable performance throughout its operation.

In pulp mills, lags obscure the effect of upstream operations on downstream measurements. Here, we estimate lags in a Canadian pulp mill using autoregressive exogenous (ARX) models. First, we show that ARX models can approximate lags in a process simulation that resembles the liquor storage tanks in pulp mills, a major source of lag in the kraft recovery cycle. Then, we use ARX models to estimate the lagged effect of a change in species pulped on as-fired liquor heating value, viscosity, and boiling point rise. Additionally, we compare the predictions of the ARX models to autoregressive (AR) models and a persistence model. The estimated lags between a change in species and heating value (49 h) and boiling point rise (41 h) agree with a detailed simulation of the mill and are close to estimated hydraulic residence times, suggesting that the liquor tanks exhibit imperfect mixing. A lagged effect of species change on viscosity could not be identified. ARX and AR models produce similar predictions that are slightly better than those of a persistence model. Finally, we show that process measurements upstream of units characterized by large residence times will likely provide little benefit to prediction accuracy.

Multi-label image classification presents a challenging task in many domains, including computer vision and medical imaging. Recent advancements have introduced graph-based and transformer-based methods to improve performance and capture label dependencies. However, these methods often include complex modules that entail heavy computation and lack interpretability. In this paper, we propose Probabilistic Multi-label Contrastive Learning (ProbMCL), a novel framework to address these challenges in multi-label image classification tasks. Our simple yet effective approach employs supervised contrastive learning, in which samples that share enough labels with an anchor image based on a decision threshold are introduced as a positive set. This structure captures label dependencies by pulling positive pair embeddings together and pushing away negative samples that fall below the threshold. We enhance representation learning by incorporating a mixture density network into contrastive learning and generating Gaussian mixture distributions to explore the epistemic uncertainty of the feature encoder. We validate the effectiveness of our framework through experimentation with datasets from the computer vision and medical imaging domains. Our method outperforms the existing state-of-the-art methods while achieving a low computational footprint on both datasets. Visualization analyses also demonstrate that ProbMCL-learned classifiers maintain a meaningful semantic topology.

This article combines the traditional definition of portfolio risk with minimum-spanning-tree-based “interconnectedness risk” to improve equal risk contribution portfolio performance. We use betweenness centrality to measure an asset’s importance in a market graph (network). After filtering the complete correlation network to a minimum spanning tree, we calculate the centrality score and convert it to a centrality heuristic. We develop an adjusted variance–covariance matrix using the centrality heuristic to bias the model to assign peripheral assets in the minimum spanning tree higher weights. We test this methodology using the constituents of the S&P 100 index. The results show that the centrality equal risk portfolio can improve upon the base equal risk portfolio returns, with a similar level of risk. We observe that during bear markets, the centrality-based portfolio can surpass the base equal risk portfolio risk.

The properties of as-fired black liquor dictate kraft recovery boiler operation. If these properties could be forecasted, operations could be adjusted to optimize boiler performance. Here, we compare the performances of classic time series models and two state-of-the-art time series neural networks for forecasting as-fired liquor heating value, viscosity, and boiling point rise at a Canadian mill. Additionally, we show that, like classic time series models, autoregressive neural networks can be regarded as functions of unknown disturbances, which is useful in comparing model complexities. Our results show that classic time series models can accurately forecast as-fired liquor properties and that classic time series models perform comparably to state-of-the-art time series neural networks. We suspect this is due to the high autocorrelation of mill data that results from frequent measurements relative to long residence times. This autocorrelation is suspected to attenuate the cross-correlations between upstream disturbances and as-fired liquor properties. As a result, neural networks, which are useful for accommodating non-linear cross-correlations and dynamics, struggle to outperform classic time series models and may not always be appropriate for forecasting chemical process parameters.

Dataset distillation aims to distill the knowledge of a large-scale real dataset into small yet informative synthetic data such that a model trained on it performs as well as a model trained on the full dataset. Despite recent progress, existing dataset distillation methods often struggle with computational efficiency, scalability to complex high-resolution datasets, and generalizability to deep architectures. These approaches typically require retraining when the distillation ratio changes, as knowledge is embedded in raw pixels. In this paper, we propose a novel framework called Data-to-Model distillation (D2M) to distill the real dataset’s knowledge into the learnable parameters of a pre-trained generative model by aligning rich representations extracted from real and generated images. The learned generative model can then produce informative training images for different distillation ratios and deep architectures. Extensive experiments on 15 datasets of varying resolutions show D2M's superior performance, re-distillation efficiency, and cross-architecture generalizability. Our method effectively scales up to high-resolution 128x128 ImageNet-1K. Furthermore, we verify D2M's practical benefits for downstream applications in neural architecture search.

Recent works in dataset distillation seek to minimize training expenses by generating a condensed synthetic dataset that encapsulates the information present in a larger real dataset. These approaches ultimately aim to attain test accuracy levels akin to those achieved by models trained on the entirety of the original dataset. Previous studies in feature and distribution matching have achieved significant results without incurring the costs of bi-level optimization in the distillation process. Despite their convincing efficiency, many of these methods suffer from marginal downstream performance improvements, limited distillation of contextual information, and subpar cross-architecture generalization. To address these challenges in dataset distillation, we propose the ATtentiOn Mixer (ATOM) module to efficiently distill large datasets using a mixture of channel and spatial-wise attention in the feature matching process. Spatial-wise attention helps guide the learning process based on consistent localization of classes in their respective images, allowing for distillation from a broader receptive field. Meanwhile, channel-wise attention captures the contextual information associated with the class itself, thus making the synthetic image more informative for training. By integrating both types of attention, our ATOM module demonstrates superior performance across various computer vision datasets, including CIFAR10/100 and Tiny-Imagenet. Notably, our method significantly improves performance in scenarios with a low number of images per class, thereby enhancing its potential. Furthermore, we maintain the improvement on cross-architectures and applications such as neural architecture search.

Generative Large Language Models enable efficient analytics across knowledge domains, rivalling human experts in information comparisons. However, the applications of LLMs for information comparisons face scalability challenges due to the difficulties in maintaining information across large contexts and overcoming model token limitations. To address these challenges, we developed the novel Abstractive Summarization Criteria-driven Comparison Endpoint (ASC^2End) system to automate information comparison at scale. Our system employs Semantic Text Similarity comparisons for generating evidence-supported analyses. We utilize proven data-handling strategies such as abstractive summarization and retrieval augmented generation to overcome token limitations and retain relevant information during model inference. Prompts were designed using zero-shot strategies to contextualize information for improved model reasoning. We evaluated abstractive summarization using ROUGE scoring and assessed the generated comparison quality using survey responses. Models evaluated on the ASC^2End system show desirable results providing insights on the expected performance of the system. ASC^2End is a novel system and tool that enables accurate, automated information comparison at scale across knowledge domains, overcoming limitations in context length and retrieval.

We examine the sensitivity of community-structured graph spectra to graph size, block size and inter-block edge probability. We use the Planted Partition Model because of its transparency. While this generative model may seem simplistic, it allows us to isolate the effects of graph and block size, edge probabilities and, consequently, vertex degree distribution on spectra. These sensitivities to key graph characteristics also generalize beyond Planted Partition Model graphs, because they are based on graph structure. Notably, our results show that eigenvalues converge to those of a complete graph, with increases in graph size or inter-block edge probability. Such convergence severely limits the use of spectral techniques.

Synthetic data, artificially generated by computer programs, has become more widely used in the financial domain to mitigate privacy concerns. Variational Autoencoder (VAE) is one of the most popular deep-learning models for generating synthetic data. However, VAE is often considered a “black box” due to its opaqueness. Although some studies have been conducted to provide explanatory insights into VAE, research focusing on explaining how the input data could influence VAE to create synthetic data, especially for tabular data, is still lacking. However, in the financial industry, most data are stored in a tabular format. This paper proposes a sensitivity-based method to assess the impact of inputted tabular data on how VAE synthesizes data. This sensitivity-based method can provide both global and local interpretations efficiently and intuitively. To test this method, a simulated dataset and three Kaggle banking tabular datasets were employed. The results confirmed the applicability of this proposed method.

This paper reviews 17 studies addressing dynamic option hedging in frictional markets through Deep Reinforcement Learning (DRL). Specifically, this work analyzes the DRL models, state and action spaces, reward formulations, data generation processes and results for each study. It is found that policy methods such as DDPG are more commonly employed due to their suitability for continuous action spaces. Despite diverse state space definitions, a lack of consensus exists on variable inclusion, prompting a call for thorough sensitivity analyses. Mean-variance metrics prevail in reward formulations, with episodic return, VaR and CvaR also yielding comparable results. Geometric Brownian motion is the primary data generation process, supplemented by stochastic volatility models like SABR (stochastic alpha, beta, rho) and the Heston model. RL agents, particularly those monitoring transaction costs, consistently outperform the Black–Scholes Delta method in frictional environments. Although consistent results emerge under constant and stochastic volatility scenarios, variations arise when employing real data. The lack of a standardized testing dataset or universal benchmark in the RL hedging space makes it difficult to compare results across different studies. A recommended future direction for this work is an implementation of DRL for hedging American options and an investigation of how DRL performs compared to other numerical American option hedging methods.

Wastewater Disinfection Using Peracids: Simulation of System Performance and Process Control Alternatives Using Batch Inactivation Kinetics from a 10-Plant North American StudyAbstractOver the last two decades, peracetic acid has gained popularity for municipal wastewater disinfection, as reported by several investigators who have conclusively demonstrated good antimicrobial properties against a wide range of microorganism models including bacteria. More recently, performic acid (PFA) has been studied as a substitute for PAA. In this study, we present recent disinfection and demand/decay and microbial inactivation kinetic results from ten North American wastewater treatment plants for PAA, PFA and total chlorine. These results are utilized in a simulated contact chamber where we consider five different dose control strategies; from simple constant disinfectant dosing to advance controls where demand and decay are measured online and adjustments to the dosing are made in real time. System performance when compared to five control strategies anticipate a lesser demand for a disinfectant for advanced control strategies at the expense of higher geomean microbial counts at the outlet. However, it is observed that using an advanced control strategy for chemical disinfection results in a steady performance and requires less quenching to meet regulatory compliance.This paper was presented at WEFTEC 2023 in Chicago, IL.SpeakerLawryshyn, YuriPresentation time16:30:0017:00:00Session time15:30:0017:00:00SessionAdvancements and Optimization with Chlorine, Peracetic Acid, and Performic Acid DisinfectionSession locationRoom S405 - Level 4TopicDisinfection and Public Health, Intermediate Level, Municipal Wastewater Treatment Design, Research and InnovationTopicDisinfection and Public Health, Intermediate Level, Municipal Wastewater Treatment Design, Research and InnovationAuthor(s)Lawryshyn, YuriAuthor(s)Y. Lawryshyn 1; L. Tikariha 1 ; J. Norton. John 2; J. Da Silva 3; K. Bell 4; D. Santoro 5; Y. Lawryshyn 1;Author affiliation(s)University of Toronto, Ontario, CANADA 1; University of Toronto, Ontario, CANADA 1 ; Great Lakes Water Authority 2; Brown and Caldwell 3; Brown and Caldwell 4; USP Technologies 5; University of Toronto 1;SourceProceedings of the Water Environment FederationDocument typeConference PaperPublisherWater Environment FederationPrint publication date Oct 2023DOI10.2175/193864718825159016Volume / Issue Content sourceWEFTECCopyright2023Word count23

This study investigates the use of a novel market graph model for equity markets. Our graph model is built on distance correlation instead of the traditional Pearson correlation. We apply it to the study of S&P500 stocks from January 2015 to December 2022. We also compare our market graphs to the traditional market graphs in the literature, those built using Pearson correlation. To further the comparison, we also build graphs using Spearman rank correlation. Our comparisons reveal that non-linear relationships in stock returns are not captured by either Pearson correlation or Spearman rank correlation. We observe that distance correlation is a robust measure for detecting complex relationships in S&P500 stock returns. Networks built on distance correlation networks, are shown to be more responsive to market conditions during turbulent periods such as the COVID crash period.

Graph isomorphism is a problem for which there is no known polynomial-time solution. The more general problem of computing graph similarity metrics, graph edit distance or maximum common subgraph, is NP-hard. Nevertheless, assessing (dis)similarity between two or more networks is a key task in many areas, such as image recognition, biology, chemistry, computer and social networks. In this article, we offer a statistical answer to the following questions: (a) "Are networks $$G_1$$ and $$G_2$$ similar?", (b) "How different are the networks $$G_1$$ and $$G_2$$ ?" and (c) "Is $$G_3$$ more similar to $$G_1$$ or $$G_2$$ ?". Our comparisons begin with the transformation of each graph into an all-pairs distance matrix. Our node-node distance, Jaccard distance, has been shown to offer an accurate reflection of the graph's connectivity structure. We then model these distances as probability distributions. Finally, we use well-established statistical tools to gauge the (dis)similarities in terms of probability distribution (dis)similarity. This comparison procedure aims to detect (dis)similarities in connectivity structure and community structure in particular, not in easily observable graph characteristics, such as degrees, edge counts or density. We validate our hypothesis that graphs can be meaningfully summarized and compared via their node-node distance distributions, using several synthetic and real-world graphs. Empirical results demonstrate its validity and the accuracy of our comparison technique.