We propose a method to learn a common bias vector for a growing sequence of low-variance tasks. Unlike state-of-the-art approaches, our method does not require tuning any hyper-parameter. Our approach is presented in the non-statistical setting and can be of two variants. The "aggressive" one updates the bias after each datapoint, the "lazy" one updates the bias only at the end of each task. We derive an across-tasks regret bound for the method. When compared to state-of-the-art approaches, the aggressive variant returns faster rates, the lazy one recovers standard rates, but with no need of tuning hyper-parameters. We then adapt the methods to the statistical setting: the aggressive variant becomes a multi-task learning method, the lazy one a meta-learning method. Experiments confirm the effectiveness of our methods in practice.

Direct loss minimization is a popular approach for learning predictors over structured label spaces. This approach is computationally appealing as it replaces integration with optimization and allows to propagate gradients in a deep net using loss-perturbed prediction. Recently, this technique was extended to generative models, while introducing a randomized predictor that samples a structure from a randomly perturbed score function. In this work, we learn the variance of these randomized structured predictors and show that it balances better between the learned score function and the randomized noise in structured prediction. We demonstrate empirically the effectiveness of learning the balance between the signal and the random noise in structured discrete spaces.

Federated learning (FL) learns a model jointly from a set of participating devices without sharing each other's privately held data. The characteristics of non-iid data across the network, low device participation, and the mandate that data remain private bring challenges in understanding the convergence of FL algorithms, particularly in regards to how convergence scales with the number of participating devices. In this paper, we focus on Federated Averaging (FedAvg)--the most widely used and effective FL algorithm in use today--and provide a comprehensive study of its convergence rate. Although FedAvg has recently been studied by an emerging line of literature, it remains open as to how FedAvg's convergence scales with the number of participating devices in the FL setting--a crucial question whose answer would shed light on the performance of FedAvg in large FL systems. We fill this gap by establishing convergence guarantees for FedAvg under three classes of problems: strongly convex smooth, convex smooth, and overparameterized strongly convex smooth problems. We show that FedAvg enjoys linear speedup in each case, although with different convergence rates. For each class, we also characterize the corresponding convergence rates for the Nesterov accelerated FedAvg algorithm in the FL setting: to the best of our knowledge, these are the first linear speedup guarantees for FedAvg when Nesterov acceleration is used. To accelerate FedAvg, we also design a new momentum-based FL algorithm that further improves the convergence rate in overparameterized linear regression problems. Empirical studies of the algorithms in various settings have supported our theoretical results.

K-Nearest Neighbors (KNN) Algoritm in Machine Learning will help you to master all the concepts of KNN. KNN algorithm is one of the simplest, non-parametric, lazy classification learning algorithm. Its purpose is to use a dataset in which the data points are separated into several classes to predict the classification of a new sample point. This tutorial on "KNN Algoritm in Machine Learning" will help you to master all the concepts of K-nearest neighbors. Its purpose is to use a dataset in which the data points are separated into several classes to predict the classification of a new sample point.

TS may look like a simple data object and easy to deal with, but the reality is that for someone new it can be a daunting task just to prepare the dataset before the actual fun stuff can begin. Every single time series (TS) data is loaded with information; and time series analysis (TSA) is the process of unpacking all of that. However, to unlock this potential, data needs to be prepared and formatted appropriately before putting it through the analytics pipeline. TS may look like a simple data object and easy to deal with, but the reality is that for someone new it can be a daunting task just to prepare the dataset before the actual fun stuff can begin. So in this article we will talk about some simple tips and tricks for getting the analysis-ready data to potentially save many hours of one's productive time.

Multiple Linear Regression with scikit-learn In this article, we studied the most fundamental machine learning algorithms i.e. linear regression. We implemented both simple linear regression and multiple linear regression with the help of the Scikit-Learn machine learning library. I hope you guys have enjoyed the reading. Let me know your doubts/suggestions in the comment section. In this 2-hour long project-based course, you will build and evaluate multiple linear regression models using Python.

In the real world, data is not as clean as it's often assumed to be. That's where all the data mining and wrangling comes in; to build insights out of the data that has been structured using queries, and now probably contains certain missing values, and exhibits possible patterns that are unseen to the naked eye. That's where Machine Learning comes in: To check for patterns and make use of those patterns to predict outcomes using these newly understood relationships in the data. For one to understand the depth of the algorithm, one needs to read through the variables in the data, and what those variables represent. Understanding this is important, because when you need to prove your outcomes, based on your understanding of the data.

Electronically excited states of molecules are at the heart of photochemistry, photophysics, as well as photobiology and also play a role in material science. Their theoretical description requires highly accurate quantum chemical calculations, which are computationally expensive. In this review, we focus on how machine learning is employed not only to speed up such excited-state simulations but also how this branch of artificial intelligence can be used to advance this exciting research field in all its aspects. Discussed applications of machine learning for excited states include excited-state dynamics simulations, static calculations of absorption spectra, as well as many others. In order to put these studies into context, we discuss the promises and pitfalls of the involved machine learning techniques. Since the latter are mostly based on quantum chemistry calculations, we also provide a short introduction into excited-state electronic structure methods, approaches for nonadiabatic dynamics simulations and describe tricks and problems when using them in machine learning for excited states of molecules.

Healthcare providers are increasingly using learned methods to predict and understand long-term patient outcomes in order to make meaningful interventions. However, despite innovations in this area, deep learning models often struggle to match performance of shallow linear models in predicting these outcomes, making it difficult to leverage such techniques in practice. In this work, motivated by the task of clinical prediction from insurance claims, we present a new technique called reverse distillation which pretrains deep models by using high-performing linear models for initialization. We make use of the longitudinal structure of insurance claims datasets to develop Self Attention with Reverse Distillation, or SARD, an architecture that utilizes a combination of contextual embedding, temporal embedding and self-attention mechanisms and most critically is trained via reverse distillation. SARD outperforms state-of-the-art methods on multiple clinical prediction outcomes, with ablation studies revealing that reverse distillation is a primary driver of these improvements.

We propose a Short-term Traffic flow Prediction (STP) framework so that transportation authorities take early actions to control flow and prevent congestion. We anticipate flow at future time frames on a target road segment based on historical flow data and innovative features such as real time feeds and trajectory data provided by Connected Vehicles (CV) technology. To cope with the fact that existing approaches do not adapt to variation in traffic, we show how this novel approach allows advanced modelling by integrating into the forecasting of flow, the impact of the various events that CV realistically encountered on segments along their trajectory. We solve the STP problem with a Deep Neural Networks (DNN) in a multitask learning setting augmented by input from CV. Results show that our approach, namely MTL-CV, with an average Root-Mean-Square Error (RMSE) of 0.052, outperforms state-of-the-art ARIMA time series (RMSE of 0.255) and baseline classifiers (RMSE of 0.122). Compared to single task learning with Artificial Neural Network (ANN), ANN had a lower performance, 0.113 for RMSE, than MTL-CV. MTL-CV learned historical similarities between segments, in contrast to using direct historical trends in the measure, because trends may not exist in the measure but do in the similarities.