Neural network binary classifiers are often evaluated on metrics like accuracy and $F_1$-Score, which are based on confusion matrix values (True Positives, False Positives, False Negatives, and True Negatives). However, these classifiers are commonly trained with a different loss, e.g. log loss. While it is preferable to perform training on the same loss as the evaluation metric, this is difficult in the case of confusion matrix based metrics because set membership is a step function without a derivative useful for backpropagation. To address this challenge, we propose an approximation of the step function that adheres to the properties necessary for effective training of binary networks using confusion matrix based metrics. This approach allows for end-to-end training of binary deep neural classifiers via batch gradient descent. We demonstrate the flexibility of this approach in several applications with varying levels of class imbalance. We also demonstrate how the approximation allows balancing between precision and recall in the appropriate ratio for the task at hand.

We investigate the benefits of heterogeneity in multi-agent explore-exploit decision making where the goal of the agents is to maximize cumulative group reward. To do so we study a class of distributed stochastic bandit problems in which agents communicate over a multi-star network and make sequential choices among options in the same uncertain environment. Typically, in multi-agent bandit problems, agents use homogeneous decision-making strategies. However, group performance can be improved by incorporating heterogeneity into the choices agents make, especially when the network graph is irregular, i.e. when agents have different numbers of neighbors. We design and analyze new heterogeneous explore-exploit strategies, using the multi-star as the model irregular network graph. The key idea is to enable center agents to do more exploring than they would do using the homogeneous strategy, as a means of providing more useful data to the peripheral agents. In the case all agents broadcast their reward values and choices to their neighbors with the same probability, we provide theoretical guarantees that group performance improves under the proposed heterogeneous strategies as compared to under homogeneous strategies. We use numerical simulations to illustrate our results and to validate our theoretical bounds.

Every nine minutes a person is diagnosed with Parkinson's Disease (PD) in the United States. However, studies have shown that between 25 and 80\% of individuals with Parkinson's Disease (PD) remain undiagnosed. An online, in the wild audio recording application has the potential to help screen for the disease if risk can be accurately assessed. In this paper, we collect data from 726 unique subjects (262 PD and 464 Non-PD) uttering the "quick brown fox jumps over the lazy dog ...." to conduct automated PD assessment. We extracted both standard acoustic features and deep learning based embedding features from the speech data and trained several machine learning algorithms on them. Our models achieved 0.75 AUC by modeling the standard acoustic features through the XGBoost model. We also provide explanation behind our model's decision and show that it is focusing mostly on the widely used MFCC features and a subset of dysphonia features previously used for detecting PD from verbal phonation task.

Multivariate time series forecasting is of great importance to many scientific disciplines and industrial sectors. The evolution of a multivariate time series depends on the dynamics of its variables and the connectivity network of causal interrelationships among them. Most of the existing time series models do not account for the causal effects among the system's variables and even if they do they rely just on determining the between-variables causality network. Knowing the structure of such a complex network and even more specifically knowing the exact lagged variables that contribute to the underlying process is crucial for the task of multivariate time series forecasting. The latter is a rather unexplored source of information to leverage. In this direction, here a novel neural network-based architecture is proposed, termed LAgged VAriable Representation NETwork (LAVARNET), which intrinsically estimates the importance of lagged variables and combines high dimensional latent representations of them to predict future values of time series. Our model is compared with other baseline and state of the art neural network architectures on one simulated data set and four real data sets from meteorology, music, solar activity, and finance areas. The proposed architecture outperforms the competitive architectures in most of the experiments.

Robots can be used to collect environmental data in regions that are difficult for humans to traverse. However, limitations remain in the size of region that a robot can directly observe per unit time. We introduce a method for selecting a limited number of observation points in a large region, from which we can predict the state of unobserved points in the region. We combine a low rank model of a target attribute with an information-maximizing path planner to predict the state of the attribute throughout a region. Our approach is agnostic to the choice of target attribute and robot monitoring platform. We evaluate our method in simulation on two real-world environment datasets, each containing observations from one to two million possible sampling locations. We compare against a random sampler and four variations of a baseline sampler from the ecology literature. Our method outperforms the baselines in terms of average Fisher information gain per samples taken and performs comparably for average reconstruction error in most trials.

Historically, intelligence is a quality that has almost always been attributed only to humans. Some scientists have advanced arguments for animal intelligence, but the consensus has always been that human intelligence is far superior than any other form of intelligence. For believers, the only intelligence greater than human intelligence is God's intelligence. For non-believers, human intelligence was the ultimate. All this was until the development of what has now become known as Artificial Intelligence (AI).

Traditional colonial power seeks unilateral power and domination over colonised people. It declares control of the social, economic, and political sphere by reordering and reinventing the social order in a manner that benefits it. In the age of algorithms, this control and domination occurs not through brute physical force but rather through invisible and nuanced mechanisms such as control of digital ecosystems and infrastructure. Common to both traditional and algorithmic colonialism is the desire to dominate, monitor, and influence the social, political, and cultural discourse through the control of core communication and infrastructure mediums. While traditional colonialism is often spearheaded by political and government forces, digital colonialism is driven by corporate tech monopolies--both of which are in search of wealth accumulation. The line between these forces is fuzzy as they intermesh and depend on one another. Political, economic, and ideological domination in the age of AI takes the form of "technological innovation", "state-of-the-art algorithms", and "AI solutions" to social problems. Algorithmic colonialism, driven by profit maximisation at any cost, assumes that the human soul, behaviour, and action is raw material free for the taking.

It's human nature to change nature. It seems part of our core program. We have the ability to imagine abstract things and then shape and mix the reality around us to create something new. As if we were gods, we have been intervening nature through history, cocreating our planet through arts, science, and the power of our fertile imagination. And when it comes to intervening our human nature itself, we don't think it twice.

We've all seen those videos of face-swapped individuals, often a parody of a popular film with a different celebrity's face superimposed over a another's via the power of artificial intelligence (AI). These videos and images have come to be known as deepfakes, and while many are indeed for harmless fun, the incredible realism of these videos leaves plenty of room for malicious behavior. Deep Trace Lab, a deepfake detection technology firm, found that the amount of detectable deepfake videos on the internet more than doubled to 49,081 in just the six months between January and June 2020. Not only that, but while deepfake prevalence was picking in the West last year, the first instances of convincing deepfakes here in Asia are starting to surface. The reported incident of face-swapping was reportedly in a Chinese TV series, when actress Liu Lu was blacklisted in the country, and her contract terminated.

Emergency response to incidents such as accidents, medical calls, and fires is one of the most pressing problems faced by communities across the globe. In the last fifty years, researchers have developed statistical, analytical, and algorithmic approaches for designing emergency response management (ERM) systems. In this survey, we present models for incident prediction, resource allocation, and dispatch for emergency incidents. We highlight the strengths and weaknesses of prior work in this domain and explore the similarities and differences between different modeling paradigms. Finally, we present future research directions. To the best of our knowledge, our work is the first comprehensive survey that explores the entirety of ERM systems.