In the domain of music and sound processing, pitch extraction plays a pivotal role. This research introduces "PitchNet", a convolutional neural network tailored for pitch extraction from the human singing voice, including acapella performances. Integrating autocorrelation with deep learning techniques, PitchNet aims to optimize the accuracy of pitch detection. Evaluation across datasets comprising synthetic sounds, opera recordings, and time-stretched vowels demonstrates its efficacy. This work paves the way for enhanced pitch extraction in both music and voice settings.
Deep reinforcement learning (DRL) is increasingly applied in large-scale productions like Netflix and Facebook. As with most data-driven systems, DRL systems can exhibit undesirable behaviors due to environmental drifts, which often occur in constantly-changing production settings. Continual Learning (CL) is the inherent self-healing approach for adapting the DRL agent in response to the environment's conditions shifts. However, successive shifts of considerable magnitude may cause the production environment to drift from its original state. Recent studies have shown that these environmental drifts tend to drive CL into long, or even unsuccessful, healing cycles, which arise from inefficiencies such as catastrophic forgetting, warm-starting failure, and slow convergence. In this paper, we propose Dr. DRL, an effective self-healing approach for DRL systems that integrates a novel mechanism of intentional forgetting into vanilla CL to overcome its main issues. Dr. DRL deliberately erases the DRL system's minor behaviors to systematically prioritize the adaptation of the key problem-solving skills. Using well-established DRL algorithms, Dr. DRL is compared with vanilla CL on various drifted environments. Dr. DRL is able to reduce, on average, the healing time and fine-tuning episodes by, respectively, 18.74% and 17.72%. Dr. DRL successfully helps agents to adapt to 19.63% of drifted environments left unsolved by vanilla CL while maintaining and even enhancing by up to 45% the obtained rewards for drifted environments that are resolved by both approaches.
Computerized Adaptive Testing (CAT) is a widely used, efficient test mode that adapts to the examinee's proficiency level in the test domain. CAT requires pre-trained item profiles, for CAT iteratively assesses the student real-time based on the registered items' profiles, and selects the next item to administer using candidate items' profiles. However, obtaining such item profiles is a costly process that involves gathering a large, dense item-response data, then training a diagnostic model on the collected data. In this paper, we explore the possibility of leveraging response data collected in the CAT service. We first show that this poses a unique challenge due to the inherent selection bias introduced by CAT, i.e., more proficient students will receive harder questions. Indeed, when naively training the diagnostic model using CAT response data, we observe that item profiles deviate significantly from the ground-truth. To tackle the selection bias issue, we propose the user-wise aggregate influence function method. Our intuition is to filter out users whose response data is heavily biased in an aggregate manner, as judged by how much perturbation the added data will introduce during parameter estimation. This way, we may enhance the performance of CAT while introducing minimal bias to the item profiles. We provide extensive experiments to demonstrate the superiority of our proposed method based on the three public datasets and one dataset that contains real-world CAT response data.
Email platforms need to generate personalized rankings of emails that satisfy user preferences, which may vary over time. We approach this as a recommendation problem based on three criteria: closeness (how relevant the sender and topic are to the user), timeliness (how recent the email is), and conciseness (how brief the email is). We propose MOSR (Multi-Objective Stationary Recommender), a novel online algorithm that uses an adaptive control model to dynamically balance these criteria and adapt to preference changes. We evaluate MOSR on the Enron Email Dataset, a large collection of real emails, and compare it with other baselines. The results show that MOSR achieves better performance, especially under non-stationary preferences, where users value different criteria more or less over time. We also test MOSR's robustness on a smaller down-sampled dataset that exhibits high variance in email characteristics, and show that it maintains stable rankings across different samples. Our work offers novel insights into how to design email re-ranking systems that account for multiple objectives impacting user satisfaction.
We propose a link acquisition time model deeply involving the process from the transmitted power to received signal-to-noise ratio (SNR) for LEO-to-ground laser communication for the first time. Compared with the conventional acquisition models founded on geometry analysis with divergence angle threshold, utilizing SNR as the decision criterion is more appropriate for practical engineering requirements. Specially, under the combined effects of platform vibration and turbulence, we decouple the parameters of beam divergence angle, spiral pitch, and coverage factor at a fixed transmitted power for a given average received SNR threshold. Then the single-scan acquisition probability is obtained by integrating the field of uncertainty (FOU), probability distribution of coverage factor, and receiver field angle. Consequently, the closed-form analytical expression of acquisition time expectation adopting multi-scan, which ensures acquisition success, with essential reset time between single-scan is derived. The optimizations concerning the beam divergence angle, spiral pitch, and FOU are presented. Moreover, the influence of platform vibration is investigated. All the analytical derivations are confirmed by Monte Carlo simulations. Notably, we provide a theoretical method for designing the minimum divergence angle modulated by the laser, which not only improves the acquisition performance within a certain vibration range, but also achieves a good trade-off with the system complexity.
The product carbon footprint (PCF) is crucial for decarbonizing the supply chain, as it measures the direct and indirect greenhouse gas emissions caused by all activities during the product's life cycle. However, PCF accounting often requires expert knowledge and significant time to construct life cycle models. In this study, we test and compare the emergent ability of five large language models (LLMs) in modeling the 'cradle-to-gate' life cycles of products and generating the inventory data of inputs and outputs, revealing their limitations as a generalized PCF knowledge database. By utilizing LLMs, we propose an automatic AI-driven PCF accounting framework, called AutoPCF, which also applies deep learning algorithms to automatically match calculation parameters, and ultimately calculate the PCF. The results of estimating the carbon footprint for three case products using the AutoPCF framework demonstrate its potential in achieving automatic modeling and estimation of PCF with a large reduction in modeling time from days to minutes.
Timely, accurate, and reliable information is essential for decision-makers, emergency managers, and infrastructure operators during flood events. This study demonstrates a proposed machine learning model, MaxFloodCast, trained on physics-based hydrodynamic simulations in Harris County, offers efficient and interpretable flood inundation depth predictions. Achieving an average R-squared of 0.949 and a Root Mean Square Error of 0.61 ft on unseen data, it proves reliable in forecasting peak flood inundation depths. Validated against Hurricane Harvey and Storm Imelda, MaxFloodCast shows the potential in supporting near-time floodplain management and emergency operations. The model's interpretability aids decision-makers in offering critical information to inform flood mitigation strategies, to prioritize areas with critical facilities and to examine how rainfall in other watersheds influences flood exposure in one area. The MaxFloodCast model enables accurate and interpretable inundation depth predictions while significantly reducing computational time, thereby supporting emergency response efforts and flood risk management more effectively.
Hurricane evacuation, ordered to save lives of people of coastal regions, generates high traffic demand with increased crash risk. To mitigate such risk, transportation agencies need to anticipate highway locations with high crash risks to deploy appropriate countermeasures. With ubiquitous sensors and communication technologies, it is now possible to retrieve micro-level vehicular data containing individual vehicle trajectory and speed information. Such high-resolution vehicle data, potentially available in real time, can be used to assess prevailing traffic safety conditions. Using vehicle speed and acceleration profiles, potential crash risks can be predicted in real time. Previous studies on real-time crash risk prediction mainly used data from infrastructure-based sensors which may not cover many road segments. In this paper, we present methods to determine potential crash risks during hurricane evacuation from an emerging alternative data source known as connected vehicle data. Such data contain vehicle location, speed, and acceleration information collected at a very high frequency (less than 30 seconds). To predict potential crash risks, we utilized a dataset collected during the evacuation period of Hurricane Ida on Interstate-10 (I-10) in the state of Louisiana. Multiple machine learning models were trained considering weather features and different traffic characteristics extracted from the connected vehicle data in 5-minute intervals. The results indicate that the Gaussian Process Boosting (GPBoost) and Extreme Gradient Boosting (XGBoost) models perform better (recall = 0.91) than other models. The real-time connected vehicle data for crash risks assessment will allow traffic managers to efficiently utilize resources to proactively take safety measures.
While recent developments in text-to-image generative models have led to a suite of high-performing methods capable of producing creative imagery from free-form text, there are several limitations. By analyzing the cross-attention representations of these models, we notice two key issues. First, for text prompts that contain multiple concepts, there is a significant amount of pixel-space overlap (i.e., same spatial regions) among pairs of different concepts. This eventually leads to the model being unable to distinguish between the two concepts and one of them being ignored in the final generation. Next, while these models attempt to capture all such concepts during the beginning of denoising (e.g., first few steps) as evidenced by cross-attention maps, this knowledge is not retained by the end of denoising (e.g., last few steps). Such loss of knowledge eventually leads to inaccurate generation outputs. To address these issues, our key innovations include two test-time attention-based loss functions that substantially improve the performance of pretrained baseline text-to-image diffusion models. First, our attention segregation loss reduces the cross-attention overlap between attention maps of different concepts in the text prompt, thereby reducing the confusion/conflict among various concepts and the eventual capture of all concepts in the generated output. Next, our attention retention loss explicitly forces text-to-image diffusion models to retain cross-attention information for all concepts across all denoising time steps, thereby leading to reduced information loss and the preservation of all concepts in the generated output.
Automatic analysis of human behaviour is a fundamental prerequisite for the creation of machines that can effectively interact with- and support humans in social interactions. In MultiMediate'23, we address two key human social behaviour analysis tasks for the first time in a controlled challenge: engagement estimation and bodily behaviour recognition in social interactions. This paper describes the MultiMediate'23 challenge and presents novel sets of annotations for both tasks. For engagement estimation we collected novel annotations on the NOvice eXpert Interaction (NOXI) database. For bodily behaviour recognition, we annotated test recordings of the MPIIGroupInteraction corpus with the BBSI annotation scheme. In addition, we present baseline results for both challenge tasks.