Nick Harris, CEO and co-founder of US photonics computing specialist Lightmatter explains how advances in photonic computing technology could give Moore's Law a shot in the arm. Recent advancements in machine learning, computer vision, natural language processing, deep learning and more are already impacting life and humanity in ways seen and often unseen. This is especially true as it relates to artificial intelligence (AI). The demands of AI are growing at a blistering rate. Training AI models today requires ultra-high performance computer chips, leading to what one might refer to as a'space race' among top technology companies to build, acquire, or get exclusive access to the highest-performance chips as soon as they come to market.

As icons and rituals adapt to newer technologies, the rise of robotics and AI can change the way we practice and experience spirituality. Some 100,000 years ago, fifteen people, eight of them children, were buried on the flank of Mount Precipice, just outside the southern edge of Nazareth in today's Israel. One of the boys still held the antlers of a large red deer clasped to his chest, while a teenager lay next to a necklace of seashells painted with ochre and brought from the Mediterranean Sea shore 35 km away. The bodies of Qafzeh are some of the earliest evidence we have of grave offerings, possibly associated with religious practice. Although some type of belief has likely accompanied us from the beginning, it's not until 50,000–13,000 BCE that we see clear religious ideas take shape in paintings, offerings, and objects. This is a period filled with Venus figurines, statuettes made of stone, bone, ivory and clay, portraying women with small heads, wide hips, and exaggerated breasts.

At the outset of World War II, you'd have a better chance of finding a needle in a haystack with a camel stuck in its eye than you did shooting down an enemy aircraft in your first dozen or so shots. This is because anti-aircraft shells at the time used manual fuses that had to be dialed in for specific lengths of time to delay their explosion. The idea was that you'd estimate where the targeted plane would be in, say five seconds, based on its currently flight path, then time the shell for that length, fire the shell at the plane and hope that the timing and location were close enough that shrapnel from the exploding shell hits the plane. If your calculations were off by even a hair, the shell would miss by thousands of feet. And if shooting down piloted aircraft was this hard, intercepting Germany's terrifyingly fast V1 and V2 rockets required far more luck than skill. But that's exactly what the team at Section T set out to do.

Sometime next April, a 50-foot-long autonomous ship will shake loose the digital bonds of its human controllers, scan the horizon with radar, and set a course westward across the Atlantic. The Mayflower Autonomous Ship won't be taking commands from a human captain like the first Mayflower did during its crossing back in 1620. Instead it will get orders from an "AI captain" built by programmers at IBM. The Mayflower's computing system processes data from 30 onboard sensors and six cameras to help the ship sail across the ocean, obey shipping rules (like how to pass other ships at sea), and control electrical and mechanical systems like the engine and rudder. There won't be anyone on board if something goes wrong, although it does have to send a daily report to a human operator back in the UK.

A key problem in computational sustainability is to understand the distribution of species across landscapes over time. This question gives rise to challenging large-scale prediction problems since (i) hundreds of species have to be simultaneously modeled and (ii) the survey data are usually inflated with zeros due to the absence of species for a large number of sites. The problem of tackling both issues simultaneously, which we refer to as the zero-inflated multi-target regression problem, has not been addressed by previous methods in statistics and machine learning. In this paper, we propose a novel deep model for the zero-inflated multi-target regression problem. To this end, we first model the joint distribution of multiple response variables as a multivariate probit model and then couple the positive outcomes with a multivariate log-normal distribution. By penalizing the difference between the two distributions' covariance matrices, a link between both distributions is established. The whole model is cast as an end-to-end learning framework and we provide an efficient learning algorithm for our model that can be fully implemented on GPUs. We show that our model outperforms the existing state-of-the-art baselines on two challenging real-world species distribution datasets concerning bird and fish populations.

The tragedy happened to the AirFrance 447 more than 10 years ago, in 2009. The flight took off in Rio de Janeiro and was planned to land in Paris. It suddenly disappeared in the middle of the Atlantic ocean without any warning. Immediately, rescuers reached the zone and what they found were just some wreckage and corpse. All 228 people onboard died in the crash.

Singular Value Decomposition (SVD) plays a pivotal role in exploratory data analysis. However, in a Big Data setting computing the dominant singular vectors is often restrictive due to the main memory requirements imposed by the dataset. Recently introduced randomized projection schemes attempt to mitigate this memory load by constructing approximate projections of the true dataset in a streaming setting. However, these projection methods come at the cost of approximation errors in both top singular values and vectors. Furthermore, in order to bound the approximation error, an over-sampled projection is required, often much larger in dimension than the desired rank. This latter consideration can still be memory intensive when the data dimension is large or extraneous when the desired rank approximation is close to the full rank. We present a two stage neural optimization approach as an alternative to conventional and randomized SVD techniques, where the memory requirement depends explicitly on the feature dimension and desired rank, independent of the sample size. The proposed scheme reads data samples in a streaming setting with the network minimization problem converging to a low rank approximation with high precision. Our architecture is fully interpretable where all the network outputs and weights have a specific meaning. We evaluate our results on various performance metrics against state of the art streaming methods. We also present numerical experiments for Singular and Eigen value decomposition on real data at various scales to show the memory efficiency of our proposed approach.

Despite being a poor and unequal country, Costa Rica has managed to close the gap in access to technology for its citizens, and it is now leading the way in the region. The country started the process of admission for the Organization for Economic Cooperation and Development (OECD) several years ago with reforms on laws, the creation of policies and the use of Computer Technologies to improve education, information access, financial markets, competitiveness, and a more open government. In May 2020, Costa Rica became the first Central American or Caribbean country invited to become an OECD member. The OECD has almost 60 years of existence, and its members are many of the world's more developed countries that work together to shape policies that foster prosperity, equality, opportunity, and well-being for their citizens. Costa Rica will become the 38th member, the fourth of Latin America.

Directional data consist of observations distributed on a (hyper)sphere, and appear in many applied fields, such as astronomy, ecology, and environmental science. This paper studies both statistical and computational problems of kernel smoothing for directional data. We generalize the classical mean shift algorithm to directional data, which allows us to identify local modes of the directional kernel density estimator (KDE). The statistical convergence rates of the directional KDE and its derivatives are derived, and the problem of mode estimation is examined. We also prove the ascending property of our directional mean shift algorithm and investigate a general problem of gradient ascent on the unit hypersphere. To demonstrate the applicability of our proposed algorithm, we evaluate it as a mode clustering method on both simulated and real-world datasets.

In the latest episode of our podcast, Machine Learning that Works, I had a great pleasure to talk to Gabriel Preda, a Lead Data Scientist at Endava and a Kaggle Grandmaster. For those of you who want to see the full interview, here is the video version. If, on the other hand, you prefer to read, I prepared a summary as well. It's not a faithful transcript of our conversation, but a structured and rephrased version of the interview, that includes the key points and observations. Without further ado, let's meet Gabriel Preda! I work for Endava, which is a software service company, and our projects are actually our clients' projects.