Three-way decision theory, which trisects the universe with less risks or costs, is considered as a powerful mathematical tool for handling uncertainty in incomplete and imprecise information tables, and provides an effective tool for conflict analysis decision making in real-time situations. In this paper, we propose the concepts of the agreement, disagreement and neutral subsets of a strategy with two evaluation functions, which establish the three-way decisions-based conflict analysis models(TWDCAMs) for trisecting the universe of agents, and employ a pair of two-way decisions models to interpret the mechanism of the three-way decision rules for an agent. Subsequently, we develop the concepts of the agreement, disagreement and neutral strategies of an agent group with two evaluation functions, which build the TWDCAMs for trisecting the universe of issues, and take a couple of two-way decisions models to explain the mechanism of the three-way decision rules for an issue. Finally, we reconstruct Fan, Qi and Wei's conflict analysis models(FQWCAMs) and Sun, Ma and Zhao's conflict analysis models(SMZCAMs) with two evaluation functions, and interpret FQWCAMs and SMZCAMs with a pair of two-day decisions models, which illustrates that FQWCAMs and SMZCAMs are special cases of TWDCAMs.

This paper proposes an approach to the joint modeling of the short-time Fourier transform magnitude and phase spectrograms with a deep generative model. We assume that the magnitude follows a Gaussian distribution and the phase follows a von Mises distribution. To improve the consistency of the phase values in the time-frequency domain, we also apply the von Mises distribution to the phase derivatives, i.e., the group delay and the instantaneous frequency. Based on these assumptions, we explore and compare several combinations of loss functions for training our models. Built upon the variational autoencoder framework, our model consists of three convolutional neural networks acting as an encoder, a magnitude decoder, and a phase decoder. In addition to the latent variables, we propose to also condition the phase estimation on the estimated magnitude. Evaluated for a time-domain speech reconstruction task, our models could generate speech with a high perceptual quality and a high intelligibility.

Reparameterizable densities are an important way to learn probability distributions in a deep learning setting. For many distributions it is possible to create low-variance gradient estimators by utilizing a `reparameterization trick'. Due to the absence of a general reparameterization trick, much research has recently been devoted to extend the number of reparameterizable distributional families. Unfortunately, this research has primarily focused on distributions defined in Euclidean space, ruling out the usage of one of the most influential class of spaces with non-trivial topologies: Lie groups. In this work we define a general framework to create reparameterizable densities on arbitrary Lie groups, and provide a detailed practitioners guide to further the ease of usage. We demonstrate how to create complex and multimodal distributions on the well known oriented group of 3D rotations, $\operatorname{SO}(3)$, using normalizing flows. Our experiments on applying such distributions in a Bayesian setting for pose estimation on objects with discrete and continuous symmetries, showcase their necessity in achieving realistic uncertainty estimates.

Interest in derivative-free optimization (DFO) and "evolutionary strategies" (ES) has recently surged in the Reinforcement Learning (RL) community, with growing evidence that they match state of the art methods for policy optimization tasks. However, blackbox DFO methods suffer from high sampling complexity since they require a substantial number of policy rollouts for reliable updates. They can also be very sensitive to noise in the rewards, actuators or the dynamics of the environment. In this paper we propose to replace the standard ES derivative-free paradigm for RL based on simple reward-weighted averaged random perturbations for policy updates, that has recently become a subject of voluminous research, by an algorithm where gradients of blackbox RL functions are estimated via regularized regression methods. In particular, we propose to use L1/L2 regularized regression-based gradient estimation to exploit sparsity and smoothness, as well as LP decoding techniques for handling adversarial stochastic and deterministic noise. Our methods can be naturally aligned with sliding trust region techniques for efficient samples reuse to further reduce sampling complexity. This is not the case for standard ES methods requiring independent sampling in each epoch. We show that our algorithms can be applied in locomotion tasks, where training is conducted in the presence of substantial noise, e.g. for learning in sim transferable stable walking behaviors for quadruped robots or training quadrupeds how to follow a path. We further demonstrate our methods on several $\mathrm{OpenAI}$ $\mathrm{Gym}$ $\mathrm{Mujoco}$ RL tasks. We manage to train effective policies even if up to $25\%$ of all measurements are arbitrarily corrupted, where standard ES methods produce sub-optimal policies or do not manage to learn at all. Our empirical results are backed by theoretical guarantees.

We study the emergence of sparse representations in neural networks. We show that in unsupervised models with regularization, the emergence of sparsity is the result of the input data samples being distributed along highly non-linear or discontinuous manifold. We also derive a similar argument for discriminatively trained networks and present experiments to support this hypothesis. Based on our study of sparsity, we introduce a new loss function which can be used as regularization term for models like autoencoders and MLPs. Further, the same loss function can also be used as a cost function for an unsupervised single-layered neural network model for learning efficient representations.

We capture millions of data points on a daily basis, a human could not process that many data points,

Artificial intelligence is finding new applications in a range of fields. Now researchers from India and Canada have developed a machine learning-based tool that can diagnose schizophrenia with high accuracy. Although research on major psychiatric illnesses has been going on for decades, there are still no reliable methods to predict and diagnose many ailments. One of reasons is the inherent variability in biological systems. Schizophrenia is a debilitating psychotic illness where diagnosis is often difficult due to its numerous clinical forms and considerable overlap with other psychiatric disorders.

If you're a company that makes robots, farm tools, security tech, or really anything that isn't a self-driving car, Waymo has a lidar to sell you. The autonomous tech company that started life in 2009 as Google's self-driving-car project announced today it's creating a new revenue stream by selling its custom-developed, short range laser sensors. It's a bit unexpected, considering Waymo waged a bruising legal fight with Uber to protect this most valuable of sensing technologies, but it also signals that Waymo is exploring business models that don't hinge on yanking the human from behind the wheel. Waymo started developing its own lidar in 2011, after deciding existing sensors--chiefly those created by Velodyne, the company that pioneered the automotive lidar market--weren't sufficient for its needs. Over the next eight years, it said during its lawsuit against Uber, Waymo put "tens of millions of dollars and tens of thousands of hours of engineering time" into its custom solution.

The power of artificial intelligence to change our lives is huge. One day, machines will be smarter than us. What are current use cases and how do they help or hurt us as humans? If planet Earth had just five pigs, three would be in China. So the spread of a deadly swine disease there is especially worrying.

Sonos has quietly updated its flagship speaker, the One, though it looks exactly the same. The Sonos One Gen 2 has been spotted by some people who have bought the new version without knowing it, but had not been announced by Sonos. Now it has become clear that it is a different version – though not one that actually changes very much. In practice, the new speaker is the same as the old one, Sonos says. But it does include a whole host of updates, even if those don't enable any new features.