Effective adaptation to distribution shifts in training data is pivotal for sustaining robustness in neural networks, especially when removing specific biases or outdated information--a process known as machine unlearning. Traditional approaches typically assume that data variations are random, which makes it difficult to adjust accurately the model parameters to remove patterns associated with the unlearned data. In this work, we present the Unlearning Information Bottleneck (UIB), a novel information-theoretic framework designed to enhance the process of machine unlearning that effectively leverages the influence of systematic patterns and biases for parameter adjustment. We propose a variational upper bound to recalibrate the model parameters through a dynamic prior that integrates changes in data distribution at an affordable computational cost, allowing efficient and accurate removal of outdated or unwanted data patterns and biases. Our experiments across various datasets, models, and unlearning methodologies demonstrate that our approach effectively removes systematic patterns and biases while maintaining the performance of models post-unlearning.

The Upper Indus Basin, Himalayas provides water for 270 million people and countless ecosystems. However, precipitation, a key component to hydrological modelling, is poorly understood in this area. A key challenge surrounding this uncertainty comes from the complex spatial-temporal distribution of precipitation across the basin. In this work we propose Gaussian processes with structured non-stationary kernels to model precipitation patterns in the UIB. Previous attempts to quantify or model precipitation in the Hindu Kush Karakoram Himalayan region have often been qualitative or include crude assumptions and simplifications which cannot be resolved at lower resolutions. This body of research also provides little to no error propagation. We account for the spatial variation in precipitation with a non-stationary Gibbs kernel parameterised with an input dependent lengthscale. This allows the posterior function samples to adapt to the varying precipitation patterns inherent in the distinct underlying topography of the Indus region. The input dependent lengthscale is governed by a latent Gaussian process with a stationary squared-exponential kernel to allow the function level hyperparameters to vary smoothly. In ablation experiments we motivate each component of the proposed kernel by demonstrating its ability to model the spatial covariance, temporal structure and joint spatio-temporal reconstruction. We benchmark our model with a stationary Gaussian process and a Deep Gaussian processes.

On February 6, 2020, an excited group of customers, partners, and investors came together to celebrate UIB's 6th anniversary. How did the UIB team celebrate this important milestone? They invited our VIP guests for an exclusive behind-the-scenes tour of UIB's new world headquarters in Ngee Ann City to experience the future of technology first-hand. Here are some of the photos from the event. Welcome to UIB! -- L-R: TGR Team Member and UIB Business Development Manager Caroline Patrick "We have entered the golden age of collaboration between humans and AI." UIB CEO Toby Ruckert welcomed everyone by saying, "We have entered the golden age of collaboration between humans and AI. Together with our partners, we are humanizing technology, not'technologizing' humans."

The Dubai Electricity and Water Authority (DEWA) recently selected UIB to be one of their partners in Dubai Future Foundation's Dubai Future Accelerators' (DFA) Cohort 6 -- UIB's third DFA cohort, our first was with du and our second was with the Dubai Police. Even though the DFA program is spread across nine weeks (starting June 16) with a four-week break in between, UIB completed the project in only three weeks, enabling DEWA to go live the week of July 8, making the project a big success for UIB, DEWA, and DFA. This has also enabled us to use the program's remaining weeks to focus on new use cases and plan for how to further scale our partnership. What did we do with DEWA? We used UIB's UnificationEngine Conversational AI platform to deploy DEWA's customer service virtual assistant, "Rammas," on WhatsApp.

The Gartner Cool Vendor logo is a trademark and service mark of Gartner, Inc., and/or its af liates, and is used herein with permission. Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner's research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or tness for a particular purpose. Attribution Gartner "Cool Vendors in Unified Communications and Collaboration Technologies" by Christopher Trueman, Gyanee Dewnarain, and Adam Preset, September 11, 2018.

The future needs to be more human (and less machine). For decades now, we have created computer programming languages and forced entire generations across the globe into becoming engineers and learning how to code. We have succeeded beyond our wildest dreams (we've created machines that can now learn on their own), and we have failed beyond our worst nightmares (we've created "black box" artificial intelligence (AI) which we don't -- and can't -- understand). It's time for us to rethink the future we're so effectively creating. I'm concerned by this trend, where we focus our mental energy into machines, rather than having them understand us more.

As published by Insights Success Magazine "Tea. You do not want to get out of bed. But the alarm on your mobile phone is ringing, and you've got things to do. So with those four iconic words, spoken into your WhatsApp, with your eyes still scrunched closed, Unified Inbox's (UIB) UnificationEngine (UE) Intelligent IoT Messaging platform hasn't just unified communications, it's turned on the smart water kettle in your kitchen. Seamlessly bringing together Artificial Intelligence (AI), the Internet of Things (IoT), and unified messaging, thanks to Natural Language Processing (NLP), those four words are boiling the water for your morning cup of tea. UIB makes it as easy for you to chat, via voice and text message, with your "things" as it is with your friends. Unified Communications company UIB's patented UE is the world's first true intelligent IoT messaging platform. UE's device- and platform-agnostic middleware enables machines and equipment to communicate with both people and things. Users can remotely control connected devices using simple natural language messages (e.g., "Tea, Earl Grey Hot" or simply, "Boil water") and receive alerts and notifications (e.g., ""Your water will be ready in two minutes" and "You have three days of milk left, shall I re-order?") on the communications channels you already use.