1

High-Throughput Training of Deep CNNs on ReRAM-based Heterogeneous Architectures via Optimized Normalization Layers

Resistive random-access memory (ReRAM)-based architectures can be used to accelerate Convolutional Neural Network (CNN) training. However, existing architectures either do not support normalization at all or they support only a limited version of it. …

Learning Pareto-Frontier Resource Management Policies for Heterogeneous SoCs - An Information-Theoretic Approach

Mobile system-on-chips (SoCs) are growing in their complexity and heterogeneity (e.g., Arm Big-Little architecture) to meet the needs of emerging applications, including games and artificial intelligence. This makes it very challenging to optimally …

Mercer Features for Efficient Combinatorial Bayesian Optimization

Bayesian optimization (BO) is an efficient framework for solving black-box optimization problems with expensive function evaluations. This paper addresses the BO problem setting for combinatorial spaces (e.g., sequences and graphs) that occurs …

Optimizing Discrete Spaces via Expensive Evaluations: A Learning to Search Framework

We consider the problem of optimizing expensive black-box functions over discrete spaces (e.g., sets, sequences, graphs). The key challenge is to select a sequence of combinatorial structures to evaluate, in order to identify high-performing …

Uncertainty-Aware Search Framework for Multi-Objective Bayesian Optimization

We consider the problem of multi-objective (MO) blackbox optimization using expensive function evaluations, where the goal is to approximate the true Pareto set of solutions while minimizing the number of function evaluations. For example, in …

Design of Multi-Output Switched-Capacitor Voltage Regulator via Machine Learning

Efficiency of power management system (PMS) is one of the key performance metrics for highly integrated system on chips (SoCs). Towards the goal of improving power efficiency of SoCs, we make two key technical contributions in this paper. First, we …

Multi-Fidelity Multi-Objective Bayesian Optimization, An Output Space Entropy Search Approach

We study the novel problem of blackbox optimization of multiple objectives via multi-fidelity function evaluations that vary in the amount of resources consumed and their accuracy. The overall goal is to approximate the true Pareto set of solutions …

Max-value Entropy Search for Multi-Objective Bayesian Optimization

We consider the problem of multi-objective (MO) blackbox optimization using expensive function evaluations, where the goal is to approximate the true Pareto-set of solutions by minimizing the number of function evaluations. For example, in hardware …

Learning and inference for structured prediction: a unifying perspective

In a structured prediction problem, one needs to learn a predictor that, given a structured input, produces a structured object, such as a sequence, tree, or clustering output. Prototypical structured prediction tasks include part-of-speech tagging …

Randomized greedy search for structured prediction: amortized inference and learning

In a structured prediction problem, we need to learn a predictor that can produce a structured output given a structured input (eg, part-of-speech tagging). The key learning and inference challenge is due to the exponential size of the structured …