Open Radio Access Network (ORAN) is being developed with an aim to democratise access and lower the cost of future mobile data networks, supporting network services with various QoS requirements, such as massive IoT and URLLC. In ORAN, network functionality is dis-aggregated into remote units (RUs), distributed units (DUs) and central units (CUs), which allows flexible software on Commercial-Off-The-Shelf (COTS) deployments. Furthermore, the mapping of variable RU requirements to local mobile edge computing centres for future centralized processing would significantly reduce the power consumption in cellular networks. In this paper, we study the RU-DU resource assignment problem in an ORAN system, modelled as a 2D bin packing problem. A deep reinforcement learning-based self-play approach is proposed to achieve efficient RU-DU resource management, with AlphaGo Zero inspired neural Monte-Carlo Tree Search (MCTS). Experiments on representative 2D bin packing environment and real sites data show that the self-play learning strategy achieves intelligent RU-DU resource assignment for different network conditions.

Fairness and explainability are two important and closely related requirements of decision making systems. While fairness and explainability of decision making systems have been extensively studied independently, only little effort has been put into studying fairness of explanations on their own. Current explanations can be unfair to an individual: an example is given by counterfactual explanations which propose different actions to change the output class to two similar individuals. In this work we formally and empirically study individual fairness and its mathematical formalization as robustness for counterfactual explanations as a prominent instance of contrasting explanations. In addition, we propose to use plausible counterfactuals instead of closest counterfactuals for improving the individual fairness of counterfactual explanations.

This article represents the second part of a series called "Letter to a CIO", which reports the discussions between the author of the letter, dr. Domenico Lepore Founder Intelligent Managemnt Inc. and several Chief Information Officers, with the aim of providing them with an effective methodology to address and successfully solve common problems that CIOs face in the Digital Age. The result of this series of interviews helped dr. A CIO MUST have the abilities necessary to accomplish the transformation from a silo-based Hierarchy to whole system optimization. Without this ability, CIOs will very soon become a relic, something that can be easily disposed of.

Batch Normalization (BN) is a commonly used technique to accelerate and stabilize training of deep neural networks. Despite its empirical success, a full theoretical understanding of BN is yet to be developed. In this work, we analyze BN through the lens of convex optimization. We introduce an analytic framework based on convex duality to obtain exact convex representations of weight-decay regularized ReLU networks with BN, which can be trained in polynomial-time. Our analyses also show that optimal layer weights can be obtained as simple closed-form formulas in the high-dimensional and/or overparameterized regimes. Furthermore, we find that Gradient Descent provides an algorithmic bias effect on the standard non-convex BN network, and we design an approach to explicitly encode this implicit regularization into the convex objective. Experiments with CIFAR image classification highlight the effectiveness of this explicit regularization for mimicking and substantially improving the performance of standard BN networks.

We propose and analyze algorithms to solve a range of learning tasks under user-level differential privacy constraints. Rather than guaranteeing only the privacy of individual samples, user-level DP protects a user's entire contribution ($m \ge 1$ samples), providing more stringent but more realistic protection against information leaks. We show that for high-dimensional mean estimation, empirical risk minimization with smooth losses, stochastic convex optimization, and learning hypothesis class with finite metric entropy, the privacy cost decreases as $O(1/\sqrt{m})$ as users provide more samples. In contrast, when increasing the number of users $n$, the privacy cost decreases at a faster $O(1/n)$ rate. We complement these results with lower bounds showing the worst-case optimality of our algorithm for mean estimation and stochastic convex optimization. Our algorithms rely on novel techniques for private mean estimation in arbitrary dimension with error scaling as the concentration radius $\tau$ of the distribution rather than the entire range. Under uniform convergence, we derive an algorithm that privately answers a sequence of $K$ adaptively chosen queries with privacy cost proportional to $\tau$, and apply it to solve the learning tasks we consider.

In this paper, we address the identical parallel machine scheduling problem with variable release dates and a common deadline for arrival. This problem occurs in several settings in which the release dates themselves are decision variables with the constraint that all jobs must arrive before or on a common fixed deadline. This deadline can be interpreted as a maximum release date for all jobs. To our knowledge, this problem has not been studied before in spite of many important applications. A first example is a manufacturing facility which uses a Just-In-Time discipline: jobs are released to the shop floor as late as possible to avoid cluttering the system but due to accounting restrictions, mostly related to the MRP (Materials Requirements Planning) logic, all work orders in a time bucket must be released before a fixed deadline. A second example is the receiving area of a warehouse which restricts the arrival of trucks within a time window. The warehouse may schedule its suppliers' trucks so to avoid congestion and provide them with an arrival time, but again, the warehouse's opening hours or external constraints such as circulation bans at certain hours, restrict the arrival of trucks. In these two examples, the deadline constraint cannot be violated, and a central controller must guarantee that all jobs meet such a constraint.

Hyperspectral imaging, also known as image spectrometry, is a landmark technique in geoscience and remote sensing (RS). In the past decade, enormous efforts have been made to process and analyze these hyperspectral (HS) products mainly by means of seasoned experts. However, with the ever-growing volume of data, the bulk of costs in manpower and material resources poses new challenges on reducing the burden of manual labor and improving efficiency. For this reason, it is, therefore, urgent to develop more intelligent and automatic approaches for various HS RS applications. Machine learning (ML) tools with convex optimization have successfully undertaken the tasks of numerous artificial intelligence (AI)-related applications. However, their ability in handling complex practical problems remains limited, particularly for HS data, due to the effects of various spectral variabilities in the process of HS imaging and the complexity and redundancy of higher dimensional HS signals. Compared to the convex models, non-convex modeling, which is capable of characterizing more complex real scenes and providing the model interpretability technically and theoretically, has been proven to be a feasible solution to reduce the gap between challenging HS vision tasks and currently advanced intelligent data processing models.

Existing deterministic variational inference approaches for diffusion processes use simple proposals and target the marginal density of the posterior. We construct the variational process as a controlled version of the prior process and approximate the posterior by a set of moment functions. In combination with moment closure, the smoothing problem is reduced to a deterministic optimal control problem. Exploiting the path-wise Fisher information, we propose an optimization procedure that corresponds to a natural gradient descent in the variational parameters. Our approach allows for richer variational approximations that extend to state-dependent diffusion terms. The classical Gaussian process approximation is recovered as a special case.

The internal structure of the model depends on the type of the There is a variety of approaches that can be used to learning algorithm, so complex data-driven models can consist identify the optimal design of the data-driven model. For of several semi-independent blocks - this approach is usually instance, AutoML solutions can be based on random search referred to as ensembling [2]. There are several techniques to [5], Bayesian optimisation [6], reinforcement learning (RL) build complex models: for example, blending allows creating [7], Monte Carlo tree search [8], sequential model-based single-level ensembles of machine learning (ML) models, and optimization [9], gradient-based approaches [10]. However, stacking allows creating multi-level ones. Other approaches are most of them are less flexible than evolutionary approaches to based on the representation of a model structure (or even the the model design (implemented e.g. in [11]). Their conceptual whole modeling pipeline) as a directed acyclic graph (DAG).

Retail is an intensely personal business, and the best artificial intelligence (AI) deployments recognize that fact. Amazon and the MIT Center for Transportation & Logistics are co-sponsoring a competition to train machine learning models to predict the delivery routes chosen by experienced drivers. Amazon is providing all information used by existing route optimization algorithms as part of the training data. However, Amazon will also provide more than 4,000 traces of driver-determined routes, which encode the drivers' know-how. Using both sources of information, contestants will be able to build models that identify and predict drivers' deviations from routes computed in the traditional manner.