Federated learning allows clients to collaboratively learn statistical models while keeping their data local. Federated learning was originally used to train a unique global model to be served to all clients, but this approach might be sub-optimal when clients' local data distributions are heterogeneous. In order to tackle this limitation, recent personalized federated learning methods train a separate model for each client while still leveraging the knowledge available at other clients. In this work, we exploit the ability of deep neural networks to extract high quality vectorial representations (embeddings) from non-tabular data, e.g., images and text, to propose a personalization mechanism based on local memorization. Personalization is obtained interpolating a pre-trained global model with a $k$-nearest neighbors (kNN) model based on the shared representation provided by the global model. We provide generalization bounds for the proposed approach and we show on a suite of federated datasets that this approach achieves significantly higher accuracy and fairness than state-of-the-art methods.

Learning from class imbalanced datasets poses challenges for many machine learning algorithms. Many real-world domains are, by definition, class imbalanced by virtue of having a majority class that naturally has many more instances than its minority class (e.g. genuine bank transactions occur much more often than fraudulent ones). Many methods have been proposed to solve the class imbalance problem, among the most popular being oversampling techniques (such as SMOTE). These methods generate synthetic instances in the minority class, to balance the dataset, performing data augmentations that improve the performance of predictive machine learning (ML) models. In this paper we advance a novel data augmentation method (adapted from eXplainable AI), that generates synthetic, counterfactual instances in the minority class. Unlike other oversampling techniques, this method adaptively combines exist-ing instances from the dataset, using actual feature-values rather than interpolating values between instances. Several experiments using four different classifiers and 25 datasets are reported, which show that this Counterfactual Augmentation method (CFA) generates useful synthetic data points in the minority class. The experiments also show that CFA is competitive with many other oversampling methods many of which are variants of SMOTE. The basis for CFAs performance is discussed, along with the conditions under which it is likely to perform better or worse in future tests.

Watson, a supercomputer by IBM, shot to fame in 2011 as the'brain' that beat two of the best contestants of Jeopardy! to win a million dollars. This system that combines artificial intelligence (AI) and sophisticated analytical software to answer questions was widely deployed in many industries. The supercomputer was developed in IBM's DeepQA project and was named after IBM's founder Thomas J. Watson. "You can be discouraged by failure, or you can learn from it. So go ahead and make mistakes, make all you can. Because, remember that's where you'll find success -- on the far side of failure."

Currently all actions created in the bot are included in the deployment version. I would like to be able, to select specific Actions, and only deploy selected Actions and not all actions in the bot. An orchestration layer managing or combining different bots might also be helpful. Within a bot, there will be various actions. You will get to a situation where you do not want to duplicate actions across bots, and use multiple bots simultaneously in one implementation.

We study the memorization power of feedforward ReLU neural networks. We show that such networks can memorize any $N$ points that satisfy a mild separability assumption using $\tilde{O}\left(\sqrt{N}\right)$ parameters. Known VC-dimension upper bounds imply that memorizing $N$ samples requires $\Omega(\sqrt{N})$ parameters, and hence our construction is optimal up to logarithmic factors. We also give a generalized construction for networks with depth bounded by $1 \leq L \leq \sqrt{N}$, for memorizing $N$ samples using $\tilde{O}(N/L)$ parameters. This bound is also optimal up to logarithmic factors. Our construction uses weights with large bit complexity. We prove that having such a large bit complexity is both necessary and sufficient for memorization with a sub-linear number of parameters.

Buy Now Pay Later (BNPL) services have increased in popularity in recent years and are ready to become a popular mode of financing. Experts claim that demand for BNPL has been accelerating in India for the past three to four years. Further, COVID-19 has boosted its demand. BNPL has now established itself as a more comfortable payment option, reducing borrowers' financial stress by providing no-cost EMIs. Uni Cards, which recently secured $18.5 million in financing, has launched its Uni Pay 1/3rd card. The product aims to enhance the customer experience in the credit card business.

Today, I continue my top-AI-stocks video series. If you are new to this series, it covers my top 12 artificial intelligence stocks focused on growth and disruptive innovation. I have done my best to find the highest-growth companies in a variety of sectors with disruptive growth trends. Last time, I shared my favorite chatbot stock. In today's video, I am covering an unknown business that Square (NYSE:SQ) acquired in 2020 that focuses on artificial intelligence and machine learning.

Square acquired this Toronto-based AI technology firm that helps combat deepfake AI cybercrime.

Today, I continue my top-AI-stocks video series.

Counterfactual explanations focus on "actionable knowledge" to help end-users understand how a machine learning outcome could be changed to a more desirable outcome. For this purpose a counterfactual explainer needs to discover input dependencies that relate to outcome changes. Identifying the minimum subset of feature changes needed to action an output change in the decision is an interesting challenge for counterfactual explainers. The DisCERN algorithm introduced in this paper is a case-based counter-factual explainer. Here counterfactuals are formed by replacing feature values from a nearest unlike neighbour (NUN) until an actionable change is observed. We show how widely adopted feature relevance-based explainers (i.e. LIME, SHAP), can inform DisCERN to identify the minimum subset of "actionable features". We demonstrate our DisCERN algorithm on five datasets in a comparative study with the widely used optimisation-based counterfactual approach DiCE. Our results demonstrate that DisCERN is an effective strategy to minimise actionable changes necessary to create good counterfactual explanations.