TY - CONF AB - Stochastic Gradient Descent (SGD) is a fundamental algorithm in machine learning, representing the optimization backbone for training several classic models, from regression to neural networks. Given the recent practical focus on distributed machine learning, significant work has been dedicated to the convergence properties of this algorithm under the inconsistent and noisy updates arising from execution in a distributed environment. However, surprisingly, the convergence properties of this classic algorithm in the standard shared-memory model are still not well-understood. In this work, we address this gap, and provide new convergence bounds for lock-free concurrent stochastic gradient descent, executing in the classic asynchronous shared memory model, against a strong adaptive adversary. Our results give improved upper and lower bounds on the "price of asynchrony'' when executing the fundamental SGD algorithm in a concurrent setting. They show that this classic optimization tool can converge faster and with a wider range of parameters than previously known under asynchronous iterations. At the same time, we exhibit a fundamental trade-off between the maximum delay in the system and the rate at which SGD can converge, which governs the set of parameters under which this algorithm can still work efficiently. AU - Alistarh, Dan-Adrian AU - De Sa, Christopher AU - Konstantinov, Nikola H ID - 5962 SN - 9781450357951 T2 - Proceedings of the 2018 ACM Symposium on Principles of Distributed Computing - PODC '18 TI - The convergence of stochastic gradient descent in asynchronous shared memory ER - TY - JOUR AB - A major problem for evolutionary theory is understanding the so-called open-ended nature of evolutionary change, from its definition to its origins. Open-ended evolution (OEE) refers to the unbounded increase in complexity that seems to characterize evolution on multiple scales. This property seems to be a characteristic feature of biological and technological evolution and is strongly tied to the generative potential associated with combinatorics, which allows the system to grow and expand their available state spaces. Interestingly, many complex systems presumably displaying OEE, from language to proteins, share a common statistical property: the presence of Zipf's Law. Given an inventory of basic items (such as words or protein domains) required to build more complex structures (sentences or proteins) Zipf's Law tells us that most of these elements are rare whereas a few of them are extremely common. Using algorithmic information theory, in this paper we provide a fundamental definition for open-endedness, which can be understood as postulates. Its statistical counterpart, based on standard Shannon information theory, has the structure of a variational problem which is shown to lead to Zipf's Law as the expected consequence of an evolutionary process displaying OEE. We further explore the problem of information conservation through an OEE process and we conclude that statistical information (standard Shannon information) is not conserved, resulting in the paradoxical situation in which the increase of information content has the effect of erasing itself. We prove that this paradox is solved if we consider non-statistical forms of information. This last result implies that standard information theory may not be a suitable theoretical framework to explore the persistence and increase of the information content in OEE systems. AU - Corominas-Murtra, Bernat AU - Seoane, Luís F. AU - Solé, Ricard ID - 5860 IS - 149 JF - Journal of the Royal Society Interface SN - 17425689 TI - Zipf's Law, unbounded complexity and open-ended evolution VL - 15 ER - TY - JOUR AB - In this paper we present a reliable method to verify the existence of loops along the uncertain trajectory of a robot, based on proprioceptive measurements only, within a bounded-error context. The loop closure detection is one of the key points in simultaneous localization and mapping (SLAM) methods, especially in homogeneous environments with difficult scenes recognitions. The proposed approach is generic and could be coupled with conventional SLAM algorithms to reliably reduce their computing burden, thus improving the localization and mapping processes in the most challenging environments such as unexplored underwater extents. To prove that a robot performed a loop whatever the uncertainties in its evolution, we employ the notion of topological degree that originates in the field of differential topology. We show that a verification tool based on the topological degree is an optimal method for proving robot loops. This is demonstrated both on datasets from real missions involving autonomous underwater vehicles and by a mathematical discussion. AU - Rohou, Simon AU - Franek, Peter AU - Aubry, Clément AU - Jaulin, Luc ID - 5960 IS - 12 JF - The International Journal of Robotics Research SN - 0278-3649 TI - Proving the existence of loops in robot trajectories VL - 37 ER - TY - CONF AB - There has been significant progress in understanding the parallelism inherent to iterative sequential algorithms: for many classic algorithms, the depth of the dependence structure is now well understood, and scheduling techniques have been developed to exploit this shallow dependence structure for efficient parallel implementations. A related, applied research strand has studied methods by which certain iterative task-based algorithms can be efficiently parallelized via relaxed concurrent priority schedulers. These allow for high concurrency when inserting and removing tasks, at the cost of executing superfluous work due to the relaxed semantics of the scheduler. In this work, we take a step towards unifying these two research directions, by showing that there exists a family of relaxed priority schedulers that can efficiently and deterministically execute classic iterative algorithms such as greedy maximal independent set (MIS) and matching. Our primary result shows that, given a randomized scheduler with an expected relaxation factor of k in terms of the maximum allowed priority inversions on a task, and any graph on n vertices, the scheduler is able to execute greedy MIS with only an additive factor of \poly(k) expected additional iterations compared to an exact (but not scalable) scheduler. This counter-intuitive result demonstrates that the overhead of relaxation when computing MIS is not dependent on the input size or structure of the input graph. Experimental results show that this overhead can be clearly offset by the gain in performance due to the highly scalable scheduler. In sum, we present an efficient method to deterministically parallelize iterative sequential algorithms, with provable runtime guarantees in terms of the number of executed tasks to completion. AU - Alistarh, Dan-Adrian AU - Brown, Trevor A AU - Kopinsky, Justin AU - Nadiradze, Giorgi ID - 5963 SN - 9781450357951 T2 - Proceedings of the 2018 ACM Symposium on Principles of Distributed Computing - PODC '18 TI - Relaxed schedulers can efficiently parallelize iterative algorithms ER - TY - CONF AB - Relaxed concurrent data structures have become increasingly popular, due to their scalability in graph processing and machine learning applications (\citeNguyen13, gonzalez2012powergraph ). Despite considerable interest, there exist families of natural, high performing randomized relaxed concurrent data structures, such as the popular MultiQueue~\citeMQ pattern for implementing relaxed priority queue data structures, for which no guarantees are known in the concurrent setting~\citeAKLN17. Our main contribution is in showing for the first time that, under a set of analytic assumptions, a family of relaxed concurrent data structures, including variants of MultiQueues, but also a new approximate counting algorithm we call the MultiCounter, provides strong probabilistic guarantees on the degree of relaxation with respect to the sequential specification, in arbitrary concurrent executions. We formalize these guarantees via a new correctness condition called distributional linearizability, tailored to concurrent implementations with randomized relaxations. Our result is based on a new analysis of an asynchronous variant of the classic power-of-two-choices load balancing algorithm, in which placement choices can be based on inconsistent, outdated information (this result may be of independent interest). We validate our results empirically, showing that the MultiCounter algorithm can implement scalable relaxed timestamps. AU - Alistarh, Dan-Adrian AU - Brown, Trevor A AU - Kopinsky, Justin AU - Li, Jerry Z. AU - Nadiradze, Giorgi ID - 5965 SN - 9781450357999 T2 - Proceedings of the 30th on Symposium on Parallelism in Algorithms and Architectures - SPAA '18 TI - Distributionally linearizable data structures ER -