IA Scholar Query: Reflective Inductive Inference of Recursive Functions.
https://scholar.archive.org/
Internet Archive Scholar query results feedeninfo@archive.orgSat, 31 Dec 2022 00:00:00 GMTfatcat-scholarhttps://scholar.archive.org/help1440Computing Graph Neural Networks: A Survey from Algorithms to Accelerators
https://scholar.archive.org/work/7uww2lnxrbdpnnyvzsanojgnba
Graph Neural Networks (GNNs) have exploded onto the machine learning scene in recent years owing to their capability to model and learn from graph-structured data. Such an ability has strong implications in a wide variety of fields whose data are inherently relational, for which conventional neural networks do not perform well. Indeed, as recent reviews can attest, research in the area of GNNs has grown rapidly and has lead to the development of a variety of GNN algorithm variants as well as to the exploration of ground-breaking applications in chemistry, neurology, electronics, or communication networks, among others. At the current stage research, however, the efficient processing of GNNs is still an open challenge for several reasons. Besides of their novelty, GNNs are hard to compute due to their dependence on the input graph, their combination of dense and very sparse operations, or the need to scale to huge graphs in some applications. In this context, this article aims to make two main contributions. On the one hand, a review of the field of GNNs is presented from the perspective of computing. This includes a brief tutorial on the GNN fundamentals, an overview of the evolution of the field in the last decade, and a summary of operations carried out in the multiple phases of different GNN algorithm variants. On the other hand, an in-depth analysis of current software and hardware acceleration schemes is provided, from which a hardware-software, graph-aware, and communication-centric vision for GNN accelerators is distilled.Sergi Abadal, Akshay Jain, Robert Guirado, Jorge López-Alonso, Eduard Alarcónwork_7uww2lnxrbdpnnyvzsanojgnbaSat, 31 Dec 2022 00:00:00 GMTRuntime Complexity Bounds Using Squeezers
https://scholar.archive.org/work/rl374t3unrghnc7chpb4a65oie
Determining upper bounds on the time complexity of a program is a fundamental problem with a variety of applications, such as performance debugging, resource certification, and compile-time optimizations. Automated techniques for cost analysis excel at bounding the resource complexity of programs that use integer values and linear arithmetic. Unfortunately, they fall short when the complexity depends more intricately on the evolution of data during execution. In such cases, state-of-the-art analyzers have shown to produce loose bounds, or even no bound at all. We propose a novel technique that generalizes the common notion of recurrence relations based on ranking functions. Existing methods usually unfold one loop iteration and examine the resulting arithmetic relations between variables. These relations assist in establishing a recurrence that bounds the number of loop iterations. We propose a different approach, where we derive recurrences by comparing whole traces with whole traces of a lower rank, avoiding the need to analyze the complexity of intermediate states. We offer a set of global properties, defined with respect to whole traces, that facilitate such a comparison and show that these properties can be checked efficiently using a handful of local conditions. To this end, we adapt state squeezers , an induction mechanism previously used for verifying safety properties. We demonstrate that this technique encompasses the reasoning power of bounded unfolding, and more. We present some seemingly innocuous, yet intricate, examples that previous tools based on cost relations and control flow analysis fail to solve, and that our squeezer-powered approach succeeds.Oren Ish-Shalom, Shachar Itzhaky, Noam Rinetzky, Sharon Shohamwork_rl374t3unrghnc7chpb4a65oieFri, 30 Sep 2022 00:00:00 GMTSession Coalgebras: A Coalgebraic View on Regular and Context-free Session Types
https://scholar.archive.org/work/7gwjvkyxsrbc5ldzvszual56sa
Compositional methods are central to the verification of software systems. For concurrent and communicating systems, compositional techniques based on behavioural type systems have received much attention. By abstracting communication protocols as types, these type systems can statically check that channels in a program interact following a certain protocol—whether messages are exchanged in the intended order. In this article, we put on our coalgebraic spectacles to investigate session types , a widely studied class of behavioural type systems. We provide a syntax-free description of session-based concurrency as states of coalgebras. As a result, we rediscover type equivalence, duality, and subtyping relations in terms of canonical coinductive presentations. In turn, this coinductive presentation enables us to derive a decidable type system with subtyping for the π-calculus, in which the states of a coalgebra will serve as channel protocols. Going full circle, we exhibit a coalgebra structure on an existing session type system, and show that the relations and type system resulting from our coalgebraic perspective coincide with existing ones. We further apply to session coalgebras the coalgebraic approach to regular languages via the so-called rational fixed point, inspired by the trinity of automata, regular languages, and regular expressions with session coalgebras, rational fixed point, and session types, respectively. We establish a suitable restriction on session coalgebras that determines a similar trinity, and reveals the mismatch between usual session types and our syntax-free coalgebraic approach. Furthermore, we extend our coalgebraic approach to account for context-free session types, by equipping session coalgebras with a stack.Alex C. Keizer, Henning Basold, Jorge A. Pérezwork_7gwjvkyxsrbc5ldzvszual56saFri, 30 Sep 2022 00:00:00 GMTNested Session Types
https://scholar.archive.org/work/cdzjx4x355eyjn7slpugmdj6di
Session types statically describe communication protocols between concurrent message-passing processes. Unfortunately, parametric polymorphism even in its restricted prenex form is not fully understood in the context of session types. In this article, we present the metatheory of session types extended with prenex polymorphism and, as a result, nested recursive datatypes. Remarkably, we prove that type equality is decidable by exhibiting a reduction to trace equivalence of deterministic first-order grammars. Recognizing the high theoretical complexity of the latter, we also propose a novel type equality algorithm and prove its soundness. We observe that the algorithm is surprisingly efficient and, despite its incompleteness, sufficient for all our examples. We have implemented our ideas by extending the Rast programming language with nested session types. We conclude with several examples illustrating the expressivity of our enhanced type system.Ankush Das, Henry Deyoung, Andreia Mordido, Frank Pfenningwork_cdzjx4x355eyjn7slpugmdj6diFri, 30 Sep 2022 00:00:00 GMTMEASURING AND ANALYZING THE IMPACT OF THE COMPONENTS OF PUBLIC SPENDING ON THE COMPONENTS OF THE TRADE BALANCE IN IRAQ FOR THE PERIOD (2004-2020)
https://scholar.archive.org/work/5aie4znqvbhw3o7pvxmei66v5y
The issue of public spending and the trade balance is one of the important topics that aroused the interest of researchers, as the research aims to measure and analyze the impact of public spending on the trade balance in Iraq for the period (2004-2020), as well as analyze the structure of public spending in terms of (current and investment) and the trade balance with both parts (exports and imports).The results of the research proved the existence of a direct and moral relationship between current spending and exports in the long term, and this means that the increase in long-term current spending by (1) billion dinars leads to an increase in exports in Iraq by (793) million dinars, while other factors remain constant. And at the level of significance (0.0041). This on the one hand agrees with the research hypothesis and on the other hand is contrary to the economic theory which says that there is an inverse relationship between current spending and exports. The results of the research proved the existence of a direct and moral relationship between current spending (CS) and imports (IM) in the long term, as the increase in current spending in the long term by (1) billion dinars leads to an increase in imports by (702) million dinars, while other factors remain fixed, and at a significant level (1%), and this agrees with the research hypothesis and the economic theory which says that there is a direct relationship between current spending and imports. The researcher's relied on the inductive approach to use modern standard models based on the Autoregressive Distributed Deceleration (ARDL) methodology, as the annual data for the period (2004-2020) was used to measure and analyze the results of the impact of public spending on the trade balance through the use of the statistical program (Eviews.9).ABDUL SATTAR SALEH MUHAMMAD AL-BILAWI, Dr. ALI AHMED DARG AL-DULAIMIwork_5aie4znqvbhw3o7pvxmei66v5yWed, 28 Sep 2022 00:00:00 GMTMathematical Components
https://scholar.archive.org/work/ahuebtxoqbcrbebz5rb2ulla4q
Mathematical Components is the name of a library of formalized mathematics for the Coq system. It covers a variety of topics, from the theory of basic data structures (e.g., numbers, lists, finite sets) to advanced results in various flavors of algebra. This library constitutes the infrastructure for the machine-checked proofs of the Four Color Theorem and of the Odd Order Theorem. The reason of existence of this book is to break down the barriers to entry. While there are several books around covering the usage of the Coq system and the theory it is based on, the Mathematical Components library is built in an unconventional way. As a consequence, this book provides a non-standard presentation of Coq, putting upfront the formalization choices and the proof style that are the pillars of the library. This books targets two classes of public. On the one hand, newcomers, even the more mathematically inclined ones, find a soft introduction to the programming language of Coq, Gallina, and the SSReflect proof language. On the other hand accustomed Coq users find a substantial account of the formalization style that made the Mathematical Components library possible.Assia Mahboubi, Enrico Tassiwork_ahuebtxoqbcrbebz5rb2ulla4qWed, 28 Sep 2022 00:00:00 GMTTowards a Verified Prover for a Ground Fragment of Set Theory
https://scholar.archive.org/work/wbn4qweybra37jlcc6f5n3l3ru
Using Isabelle/HOL, we verify the state-of-the-art decision procedure for multi-level syllogistic with singleton (MLSS for short), which is a ground fragment of set theory. We formalise the syntax and semantics of MLSS as well as a sound and complete tableau calculus for it. We also provide an abstract specification of a decision procedure that applies the rules of the calculus exhaustively and prove its termination.Lukas Stevenswork_wbn4qweybra37jlcc6f5n3l3ruWed, 28 Sep 2022 00:00:00 GMTMechanizing the comparison of the semantic expressiveness of recursive types
https://scholar.archive.org/work/qnwunkjrivaf7b6nqitdaipice
Recursive types extend the simply-typed lambda calculus (STLC) with the additional expressive power to enable diverging computation and to encode recursive data-types (e.g., lists). Two formulations of recursive types exist: iso-recursive and equi-recursive. The relative advantages of iso- and equi-recursion are well-studied when it comes to their impact on type-inference. However, the relative semantic expressiveness of the two formulations remains unclear so far. This paper studies the semantic expressiveness of STLC with iso- and equi-recursive types, proving that these formulations are equally expressive. In fact, we prove that they are both as expressive as STLC with only term-level recursion. We phrase these equi-expressiveness results in terms of full abstraction of three canonical compilers between these three languages (STLC with iso-, with equi-recursive types and with term-level recursion). Our choice of languages allows us to study expressiveness when interacting over both a simply-typed and a recursively-typed interface. The three proofs all rely on a typed version of a proof technique called approximate backtranslation. Together, our results show that there is no difference in semantic expressiveness between STLCs with iso- and equi-recursive types. In this paper, we focus on a simply-typed setting but we believe our results scale to more powerful type systems like System F.Dominique Devriese, Eric Mark Martin, Marco Patrignaniwork_qnwunkjrivaf7b6nqitdaipiceWed, 28 Sep 2022 00:00:00 GMTThe Theory of Duality and Periodicity
https://scholar.archive.org/work/53ay35j3dbdnlotkppxfr552le
Dualism is a metaphysical, philosophical concept which refers to two irreducible, heterogeneous principles. This idea is known to appear in a lot of places in the universe, however a rigorous mathematical definition and theory is not yet established in a formal way. In this paper, we develop a novel theory to represent philosophical dualism in a formal mathematical construction with the context of quantum physics, known as the "theory of duality". We will use traditional Chinese philosophical concepts in duality as the foundation as it greatly resembles to the mathematical and physical construction for our purpose. The idea of periodicity based on Taoism will also be introduced mathematically. This paper will demonstrate how to convolve metaphysical idea into mathematics and physics. Finally, we will implement the concept of duality to prove some fundamental theorems of Buddhism.B.T.T.Wongwork_53ay35j3dbdnlotkppxfr552leTue, 27 Sep 2022 00:00:00 GMTEmbedding Hindsight Reasoning in Separation Logic
https://scholar.archive.org/work/7llk2wfmkfekzegg5iqmfsfp2i
Proving linearizability of concurrent data structures remains a key challenge for verification. We present temporal interpolation as a new proof principle to conduct such proofs using hindsight arguments within concurrent separation logic. Temporal reasoning offers an easy-to-use alternative to prophecy variables and has the advantage of structuring proofs into easy-to-discharge hypotheses. To hindsight theory, our work brings the formal rigor and proof machinery of concurrent program logics. We substantiate the usefulness of our development by verifying the linearizability of the Logical Ordering (LO-)tree and RDCSS. Both of these involve complex proof arguments due to future-dependent linearization points. The LO-tree additionally features complex structure overlays. Our proof of the LO-tree is the first formal proof of this data structure. Interestingly, our formalization revealed an unknown bug and an existing informal proof as erroneous.Roland Meyer, Thomas Wies, Sebastian Wolffwork_7llk2wfmkfekzegg5iqmfsfp2iTue, 27 Sep 2022 00:00:00 GMTHigh-Dimensional Geometric Streaming in Polynomial Space
https://scholar.archive.org/work/btmazlynqvc77kwthohpbovcia
Many existing algorithms for streaming geometric data analysis have been plagued by exponential dependencies in the space complexity, which are undesirable for processing high-dimensional data sets. In particular, once d≥log n, there are no known non-trivial streaming algorithms for problems such as maintaining convex hulls and Löwner-John ellipsoids of n points, despite a long line of work in streaming computational geometry since [AHV04]. We simultaneously improve these results to poly(d,log n) bits of space by trading off with a poly(d,log n) factor distortion. We achieve these results in a unified manner, by designing the first streaming algorithm for maintaining a coreset for ℓ_∞ subspace embeddings with poly(d,log n) space and poly(d,log n) distortion. Our algorithm also gives similar guarantees in the online coreset model. Along the way, we sharpen results for online numerical linear algebra by replacing a log condition number dependence with a log n dependence, answering a question of [BDM+20]. Our techniques provide a novel connection between leverage scores, a fundamental object in numerical linear algebra, and computational geometry. For ℓ_p subspace embeddings, we give nearly optimal trade-offs between space and distortion for one-pass streaming algorithms. For instance, we give a deterministic coreset using O(d^2log n) space and O((dlog n)^1/2-1/p) distortion for p>2, whereas previous deterministic algorithms incurred a poly(n) factor in the space or the distortion [CDW18]. Our techniques have implications in the offline setting, where we give optimal trade-offs between the space complexity and distortion of subspace sketch data structures. To do this, we give an elementary proof of a "change of density" theorem of [LT80] and make it algorithmic.David P. Woodruff, Taisuke Yasudawork_btmazlynqvc77kwthohpbovciaTue, 27 Sep 2022 00:00:00 GMTDescriptive vs. inferential community detection in networks: pitfalls, myths, and half-truths
https://scholar.archive.org/work/7kx6vshwkjabhnaukqodb5j6ka
Community detection is one of the most important methodological fields of network science, and one which has attracted a significant amount of attention over the past decades. This area deals with the automated division of a network into fundamental building blocks, with the objective of providing a summary of its large-scale structure. Despite its importance and widespread adoption, there is a noticeable gap between what is arguably the state-of-the-art and the methods that are actually used in practice in a variety of fields. Here we attempt to address this discrepancy by dividing existing methods according to whether they have a "descriptive" or an "inferential" goal. While descriptive methods find patterns in networks based on context-dependent notions of community structure, inferential methods articulate generative models, and attempt to fit them to data. In this way, they are able to provide insights into the mechanisms of network formation, and separate structure from randomness in a manner supported by statistical evidence. We review how employing descriptive methods with inferential aims is riddled with pitfalls and misleading answers, and thus should be in general avoided. We argue that inferential methods are more typically aligned with clearer scientific questions, yield more robust results, and should be in many cases preferred. We attempt to dispel some myths and half-truths often believed when community detection is employed in practice, in an effort to improve both the use of such methods as well as the interpretation of their results.Tiago P. Peixotowork_7kx6vshwkjabhnaukqodb5j6kaMon, 26 Sep 2022 00:00:00 GMTSelf-Adaptive Forecasting for Improved Deep Learning on Non-Stationary Time-Series
https://scholar.archive.org/work/2d6g5lenknhrrnjmayajhp7cs4
Real-world time-series datasets often violate the assumptions of standard supervised learning for forecasting -- their distributions evolve over time, rendering the conventional training and model selection procedures suboptimal. In this paper, we propose a novel method, Self-Adaptive Forecasting (SAF), to modify the training of time-series forecasting models to improve their performance on forecasting tasks with such non-stationary time-series data. SAF integrates a self-adaptation stage prior to forecasting based on 'backcasting', i.e. predicting masked inputs backward in time. This is a form of test-time training that creates a self-supervised learning problem on test samples before performing the prediction task. In this way, our method enables efficient adaptation of encoded representations to evolving distributions, leading to superior generalization. SAF can be integrated with any canonical encoder-decoder based time-series architecture such as recurrent neural networks or attention-based architectures. On synthetic and real-world datasets in domains where time-series data are known to be notoriously non-stationary, such as healthcare and finance, we demonstrate a significant benefit of SAF in improving forecasting accuracy.Sercan O. Arik, Nathanael C. Yoder, Tomas Pfisterwork_2d6g5lenknhrrnjmayajhp7cs4Mon, 26 Sep 2022 00:00:00 GMTProtoPFormer: Concentrating on Prototypical Parts in Vision Transformers for Interpretable Image Recognition
https://scholar.archive.org/work/2kfhihvtcja4pn47jhojmrtk4q
Prototypical part network (ProtoPNet) has drawn wide attention and boosted many follow-up studies due to its self-explanatory property for explainable artificial intelligence (XAI). However, when directly applying ProtoPNet on vision transformer (ViT) backbones, learned prototypes have a "distraction" problem: they have a relatively high probability of being activated by the background and pay less attention to the foreground. The powerful capability of modeling long-term dependency makes the transformer-based ProtoPNet hard to focus on prototypical parts, thus severely impairing its inherent interpretability. This paper proposes prototypical part transformer (ProtoPFormer) for appropriately and effectively applying the prototype-based method with ViTs for interpretable image recognition. The proposed method introduces global and local prototypes for capturing and highlighting the representative holistic and partial features of targets according to the architectural characteristics of ViTs. The global prototypes are adopted to provide the global view of objects to guide local prototypes to concentrate on the foreground while eliminating the influence of the background. Afterwards, local prototypes are explicitly supervised to concentrate on their respective prototypical visual parts, increasing the overall interpretability. Extensive experiments demonstrate that our proposed global and local prototypes can mutually correct each other and jointly make final decisions, which faithfully and transparently reason the decision-making processes associatively from the whole and local perspectives, respectively. Moreover, ProtoPFormer consistently achieves superior performance and visualization results over the state-of-the-art (SOTA) prototype-based baselines. Our code has been released at https://github.com/zju-vipa/ProtoPFormer.Mengqi Xue, Qihan Huang, Haofei Zhang, Lechao Cheng, Jie Song, Minghui Wu, Mingli Songwork_2kfhihvtcja4pn47jhojmrtk4qMon, 26 Sep 2022 00:00:00 GMTDetecting quantumness in uniform precessions
https://scholar.archive.org/work/dampma32qncujksbyra6qgvruu
Building on work by Tsirelson, we present a family of protocols that detect the nonclassicality of suitable states of a single quantum system, under the sole assumption that the measured dynamical observable undergoes a uniform precession. The case of the harmonic oscillator was anticipated in the work by Tsirelson, which we extend. We then apply the protocols to finite-dimensional spins that undergo uniform precession in real space and find a gap between the classical and the quantum expectations for every j≥3/2 (excluding j=2).Lin Htoo Zaw, Clive Cenxin Aw, Zakarya Lasmar, Valerio Scaraniwork_dampma32qncujksbyra6qgvruuMon, 26 Sep 2022 00:00:00 GMTNormal functions and maximal order types
https://scholar.archive.org/work/7ut6mggqirhnrluqcnnkd3s4ey
Transformations of well partial orders induce functions on the ordinals, via the notion of maximal order type. In most examples from the literature, these functions are not normal, in marked contrast with the central role that normal functions play in ordinal analysis and related work from computability theory. The present paper aims to explain this phenomenon. In order to do so, we investigate a rich class of order transformations that are known as 𝖶𝖯𝖮-dilators. According to a first main result of this paper, 𝖶𝖯𝖮-dilators induce normal functions when they satisfy a rather restrictive condition, which we call strong normality. Moreover, the reverse implication holds as well, for reasonably well behaved 𝖶𝖯𝖮-dilators. Strong normality also allows us to explain another phenomenon: by previous work of Freund, Rathjen and Weiermann, a uniform Kruskal theorem for 𝖶𝖯𝖮-dilators is as strong as Π^1_1-comprehension, while the corresponding result for normal dilators on linear orders is equivalent to the much weaker principle of Π^1_1-induction. As our second main result, we show that Π^1_1-induction is equivalent to the uniform Kruskal theorem for 𝖶𝖯𝖮-dilators that are strongly normal.Anton Freund, Davide Mancawork_7ut6mggqirhnrluqcnnkd3s4eyFri, 23 Sep 2022 00:00:00 GMTLearning Interpretable Dynamics from Images of a Freely Rotating 3D Rigid Body
https://scholar.archive.org/work/iyitxlg2qjb2doiwa43ydxakbq
In many real-world settings, image observations of freely rotating 3D rigid bodies, such as satellites, may be available when low-dimensional measurements are not. However, the high-dimensionality of image data precludes the use of classical estimation techniques to learn the dynamics and a lack of interpretability reduces the usefulness of standard deep learning methods. In this work, we present a physics-informed neural network model to estimate and predict 3D rotational dynamics from image sequences. We achieve this using a multi-stage prediction pipeline that maps individual images to a latent representation homeomorphic to 𝐒𝐎(3), computes angular velocities from latent pairs, and predicts future latent states using the Hamiltonian equations of motion with a learned representation of the Hamiltonian. We demonstrate the efficacy of our approach on a new rotating rigid-body dataset with sequences of rotating cubes and rectangular prisms with uniform and non-uniform density.Justice Mason and Christine Allen-Blanchette and Nicholas Zolman and Elizabeth Davison and Naomi Leonardwork_iyitxlg2qjb2doiwa43ydxakbqFri, 23 Sep 2022 00:00:00 GMTExpanding the Deployment Envelope of Behavior Prediction via Adaptive Meta-Learning
https://scholar.archive.org/work/qbfj5msyrrgxljt5e654g23wea
Learning-based behavior prediction methods are increasingly being deployed in real-world autonomous systems, e.g., in fleets of self-driving vehicles, which are beginning to commercially operate in major cities across the world. Despite their advancements, however, the vast majority of prediction systems are specialized to a set of well-explored geographic regions or operational design domains, complicating deployment to additional cities, countries, or continents. Towards this end, we present a novel method for efficiently adapting behavior prediction models to new environments. Our approach leverages recent advances in meta-learning, specifically Bayesian regression, to augment existing behavior prediction models with an adaptive layer that enables efficient domain transfer via offline fine-tuning, online adaptation, or both. Experiments across multiple real-world datasets demonstrate that our method can efficiently adapt to a variety of unseen environments.Boris Ivanovic, James Harrison, Marco Pavonework_qbfj5msyrrgxljt5e654g23weaFri, 23 Sep 2022 00:00:00 GMTMagnetic interactions in orbital dynamics
https://scholar.archive.org/work/wek2impms5f47ldtaau24umpia
The magnetic field of a host star can impact the orbit of a stellar partner, planet, or asteroid if the orbiting body is itself magnetic or electrically conducting. Here, we focus on the instantaneous magnetic forces on an orbiting body in the limit where the dipole approximation describes its magnetic properties as well as those of its stellar host. A permanent magnet in orbit about a star will be inexorably drawn toward the stellar host if the magnetic force is comparable to gravity due to the steep radial dependence of the dipole-dipole interaction. While magnetic fields in observed systems are much too weak to drive a merger event, we confirm that they may be high enough in some close compact binaries to cause measurable orbital precession. When the orbiting body is a conductor, the stellar field induces a time-varying magnetic dipole moment that leads to the possibility of eccentricity pumping and resonance trapping. The challenge is that the orbiter must be close to the stellar host, so that magnetic interactions must compete with tidal forces and the effects of intense stellar radiation.Benjamin C. Bromley, Scott J. Kenyonwork_wek2impms5f47ldtaau24umpiaFri, 23 Sep 2022 00:00:00 GMTCombinatorial optimization and reasoning with graph neural networks
https://scholar.archive.org/work/dszclpgdgfgzrnd562tfbceni4
Combinatorial optimization is a well-established area in operations research and computer science. Until recently, its methods have focused on solving problem instances in isolation, ignoring that they often stem from related data distributions in practice. However, recent years have seen a surge of interest in using machine learning, especially graph neural networks (GNNs), as a key building block for combinatorial tasks, either directly as solvers or by enhancing exact solvers. The inductive bias of GNNs effectively encodes combinatorial and relational input due to their invariance to permutations and awareness of input sparsity. This paper presents a conceptual review of recent key advancements in this emerging field, aiming at optimization and machine learning researchers.Quentin Cappart, Didier Chételat, Elias Khalil, Andrea Lodi, Christopher Morris, Petar Veličkovićwork_dszclpgdgfgzrnd562tfbceni4Fri, 23 Sep 2022 00:00:00 GMT