AI Problems Research Articles

Backdoors to planning

Backdoors measure the distance to tractable fragments and have become an important tool to find fixed-parameter tractable (fpt) algorithms for hard problems in AI and beyond. Despite their success, backdoors have not been used for planning, a central problem in AI that has a high computational complexity. In this work, we introduce two notions of backdoors building upon the causal graph. We analyze the complexity of finding a small backdoor (detection) and using the backdoor to solve the problem (evaluation) in the light of planning with (un)bounded plan length/domain of the variables. For each setting we present either an fpt-result or rule out the existence thereof by showing parameterized intractability. For several interesting cases we achieve the most desirable outcome: detection and evaluation are fpt. In addition, we explore the power of polynomial preprocessing for all fpt-results, i.e., we investigate whether polynomial kernels exist. We show that for the detection problems, polynomial kernels exist whereas we rule out the existence of polynomial kernels for the evaluation problems.

IBM Goes Full Steam toward Eliminating AI Problems

IBM Goes Full Steam toward Eliminating AI Problems

Successful Nash Equilibrium Agent for a Three-Player Imperfect-Information Game

Creating strong agents for games with more than two players is a major open problem in AI. Common approaches are based on approximating game-theoretic solution concepts such as Nash equilibrium, which have strong theoretical guarantees in two-player zero-sum games, but no guarantees in non-zero-sum games or in games with more than two players. We describe an agent that is able to defeat a variety of realistic opponents using an exact Nash equilibrium strategy in a three-player imperfect-information game. This shows that, despite a lack of theoretical guarantees, agents based on Nash equilibrium strategies can be successful in multiplayer games after all.

Machine Theorem Discovery

This article describes a framework for machine theorem discovery and illustrates its use in discovering state invariants in planning domains and properties about Nash equilibria in game theory. It also discusses its potential use in program verification in software engineering. The main message of the article is that many AI problems can and should be formulated as machine theorem discovery tasks.

An Exhaustive DPLL Algorithm for Model Counting

State-of-the-art model counters are based on exhaustive DPLL algorithms, and have been successfully used in probabilistic reasoning, one of the key problems in AI. In this article, we present a new exhaustive DPLL algorithm with a formal semantics, a proof of correctness, and a modular design. The modular design is based on the separation of the core model counting algorithm from SAT solving techniques. We also show that the trace of our algorithm belongs to the language of Sentential Decision Diagrams (SDDs), which is a subset of Decision-DNNFs, the trace of existing state-of-the-art model counters. Still, our experimental analysis shows comparable results against state-of-the-art model counters. Furthermore, we obtain the first top-down SDD compiler, and show orders-of-magnitude improvements in SDD construction time against the existing bottom-up SDD compiler.

HogRider: Champion Agent of Microsoft Malmo Collaborative AI Challenge

It has been an open challenge for self-interested agents to make optimal sequential decisions in complex multiagent systems, where agents might achieve higher utility via collaboration. The Microsoft Malmo Collaborative AI Challenge (MCAC), which is designed to encourage research relating to various problems in Collaborative AI, takes the form of a Minecraft mini-game where players might work together to catch a pig or deviate from cooperation, for pursuing high scores to win the challenge. Various characteristics, such as complex interactions among agents, uncertainties, sequential decision making and limited learning trials all make it extremely challenging to find effective strategies. We present HogRider---the champion agent of MCAC in 2017 out of 81 teams from 26 countries. One key innovation of HogRider is a generalized agent type hypothesis framework to identify the behavior model of the other agents, which is demonstrated to be robust to observation uncertainty. On top of that, a second key innovation is a novel Q-learning approach to learn effective policies against each type of the collaborating agents. Various ideas are proposed to adapt traditional Q-learning to handle complexities in the challenge, including state-action abstraction to reduce problem scale, a warm start approach using human reasoning for addressing limited learning trials, and an active greedy strategy to balance exploitation-exploration. Challenge results show that HogRider outperforms all the other teams by a significant edge, in terms of both optimality and stability.

Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations

Human action recognition remains a difficult problem for AI. Traditional machine learning techniques can have high recognition accuracy, but they are typically black boxes whose internal models are not inspectable and whose results are not explainable. This paper describes a new pipeline for recognizing human actions from skeleton data via analogical generalization. Specifically, starting with Kinect data, we segment each human action by temporal regions where the motion is qualitatively uniform, creating a sketch graph that provides a form of qualitative representation of the behavior that is easy to visualize. Models are learned from sketch graphs via analogical generalization, which are then used for classification via analogical retrieval. The retrieval process also produces links between the new example and components of the model that provide explanations. To improve recognition accuracy, we implement dynamic feature selection to pick reasonable relational features. We show the explanation advantage of our approach by example, and results on three public datasets illustrate its utility.

Phase‐Mapper: Accelerating Materials Discovery with AI

From the stone age to the bronze, iron, and modern silicon ages, the discovery and characterization of new materials has always been instrumental to humanity's development and progress. With the current pressing need to address sustainability challenges and find alternatives to fossil fuels, we look for solutions in the development of new materials that will allow for renewable energy. To discover materials with the required properties, materials scientists can perform high‐throughput materials discovery, which includes rapid synthesis and characterization via X‐ray diffraction (XRD) of thousands of materials. A central problem in materials discovery, the phase map identification problem, involves the determination of the crystal structure of materials from materials composition and structural characterization data. This analysis is traditionally performed mainly by hand, which can take days for a single material system. In this work we present Phase‐Mapper, a solution platform that tightly integrates XRD experimentation, AI problem solving, and human intelligence for interpreting XRD patterns and inferring the crystal structures of the underlying materials. Phase‐Mapper is compatible with any spectral demixing algorithm, including our novel solver, AgileFD, which is based on convolutive nonnegative matrix factorization (NMF). AgileFD allows materials scientists to rapidly interpret XRD patterns, and incorporates constraints to capture prior knowledge about the physics of the materials as well as human feedback. With our system, materials scientists have been able to interpret previously unsolvable systems of XRD data at the Department of Energy's Joint Center for Artificial Photosynthesis, including the Nb‐Mn‐V oxide system, which led to the discovery of new solar light absorbers and is provided as an illustrative example of AI‐enabled high‐throughput materials discovery.

The Race for an Artificial General Intelligence: Implications for Public Policy

An arms race for an artificial general intelligence (AGI) would be detrimental for and even pose an existential threat to humanity if it results in an unfriendly AGI. In this paper an all-pay contest model is developed to derive implications for public policy to avoid such an outcome. It is established that in a winner-takes all race, where players must invest in R&D, only the most competitive teams will participate. Given the difficulty of AGI the number of competing teams is unlikely ever to be very large. It is also established that the intention of teams competing in an AGI race, as well as the possibility of an intermediate prize is important in determining the quality of the eventual AGI. The possibility of an intermediate prize will raise quality of research but also the probability of finding the dominant AGI application and hence will make public control more urgent. It is recommended that the danger of an unfriendly AGI can be reduced by taxing AI and by using public procurement. This would reduce the pay-off of contestants, raise the amount of R&D needed to compete, and coordinate and incentivize co-operation, all outcomes that will help alleviate the control and political problems in AI. Future research is needed to elaborate the design of systems of public procurement of AI innovation and for appropriately adjusting the legal frameworks underpinning high-tech innovation, in particular dealing with patents created by AI.

The Race for an Artificial General Intelligence: Implications for Public Policy

An arms race for an artificial general intelligence (AGI) would be detrimental for and even pose an existential threat to humanity if it results in an unfriendly AGI. In this paper an all-pay contest model is developed to derive implications for public policy to avoid such an outcome. It is established that in a winner-takes all race, where players must invest in R&D, only the most competitive teams will participate. Given the difficulty of AGI the number of competing teams is unlikely ever to be very large. It is also established that the intention of teams competing in an AGI race, as well as the possibility of an intermediate prize is important in determining the quality of the eventual AGI.The possibility of an intermediate prize will raise quality of research but also the probability of finding the dominant AGI application and hence will make public control more urgent. It is recommended that the danger of an unfriendly AGI can be reduced by taxing AI and by using public procurement. This would reduce the pay-off of contestants, raise the amount of R&D needed to compete, and coordinate and incentivize co-operation, all outcomes that will help alleviate the control and political problems in AI. Future research is needed to elaborate the design of systems of public procurement of AI innovation and for appropriately adjusting the legal frameworks underpinning high-tech innovation, in particular dealing with patents created by AI.

Reading the Moving Text in Animated Text-Based CAPTCHAs

Having based on hard AI problems, CAPTCHA (Completely Automated Public Turing test to tell the Computers and Humans Apart) is a hot research topic in the field of computer vision and artificial intelligence. CAPTCHA is a challenge-response test conducted to single out humans and bots. It is ubiquitously implemented on the web since its introduction. As text-based CAPTCHAs are successfully broken by various researchers therefore several design variants have been proposed and implemented in order to further strengthen it. Animated Text-based CAPTCHAs are one of the design variant of it and are based on the difficulty of reading the moving text. They are based on zero knowledge per frame principle. Although it’s still easy for humans to read animated text but it’s a challenge for machines. As proposals for animated CAPTCHAs are on the rise so there is a strong need to scrutinize their strength against automated attacks. In this research, such CAPTCHAs are investigated to verify their robustness against automated attacks. The proposed methods proved that these CAPTCHAs are vulnerable and they do not guarantee the robustness against automated attacks. The proposed frame selection, noise removal, segmentation and recognition methods have successfully decoded these CAPTCHAs with an overall precision, segmentation accuracy and recognition rate of up to 53.8%, 92.9% and 93.5% respectively.

No Bot Expects the DeepCAPTCHA! Introducing Immutable Adversarial Examples, With Applications to CAPTCHA Generation

Recent advances in deep learning (DL) allow for solving complex AI problems that used to be considered very hard. While this progress has advanced many fields, it is considered to be bad news for Completely Automated Public Turing tests to tell Computers and Humans Apart (CAPTCHAs), the security of which rests on the hardness of some learning problems. In this paper, we introduce DeepCAPTCHA, a new and secure CAPTCHA scheme based on adversarial examples , an inherit limitation of the current DL networks. These adversarial examples are constructed inputs, either synthesized from scratch or computed by adding a small and specific perturbation called adversarial noise to correctly classified items, causing the targeted DL network to misclassify them. We show that plain adversarial noise is insufficient to achieve secure CAPTCHA schemes, which leads us to introduce immutable adversarial noise —an adversarial noise that is resistant to removal attempts. In this paper, we implement a proof of concept system, and its analysis shows that the scheme offers high security and good usability compared with the best previously existing CAPTCHAs.

Assessment on problems associated with artificial insemination service in dairy cattle in Tullo district, West Hararghe, Ethiopia

A cross sectional study was conducted from November 2015 to May 2016 with the objective of assessing the problems associated with artificial insemination service in dairy cattle in five selected peasant associations of Tullo district, west Hararghe, Ethiopia. A structured questionnaire was used for 389 dairy cattle owners, 3 artificial insemination technicians and 8 animal health professionals. The result of the study showed that 219 (56.3%) of the dairy farmers get artificial insemination services regularly and without interruption while 170 (43.7%) do not use this service regularly due to shortage of artificial insemination technicians (18.2%), discontinuation of service on weekends and holidays (51.2%) and shortage of inputs (30.6.0%). Conception failure (34.9%) was identified by dairy cattle owners as a major problem of AI followed by death of embryo or dystocia (28%). About 71.1% of the owners wait for the next oestrus as they are unable to get AI service during heat period while 28.9% use natural mating if they do not get the service. The questionnaire survey indicated that artificial insemination service faces several constraints and needs improvement for increasing the productivity of dairy cattle in term of genetic improvement in the study area. Keywords: Artificial insemination; Cross breed; Cross sectional; Dairy cattle; Tullo

포커 게임에서의 인공지능의 현실과 문제점

포커 게임에서의 인공지능의 현실과 문제점

Compressed Path Databases with Ordered Wildcard Substitutions

Compressed path databases (CPDs) are a state-of-the-art approach to path planning, a core AI problem. In the Grid-based Path Planning Competition, the CPD-based SRC path planning system was the fastest competitor with respect to both computing full optimal paths and computing the first moves of an optimal path. However, on large maps, CPDs can require a significant amount of memory, which can be a serious practical bottleneck. We present an approach that significantly reduces the size of a CPD. Our approach replaces part of the data encoded in a CPD with wildcards ("don’t care" symbols), maintaining the ability to compute optimal paths for all pairs of nodes of an undirected graph. We show that using wildcards in a way that maximizes the memory savings is NP-hard. We consider heuristics that achieve a good performance in practice. We implement our ideas on top of SRC and provide a detailed empirical analysis. Average memory savings can reach a factor of 2. Our first-k-moves lag (i.e., the time before knowing the first k optimal forward moves) increases, but it can be kept within competitive values. The speed of computing full optimal paths improves slightly.

ディープラーニングと人工知能の難問( 深化する機械学習-技術の進展とその応用)

ディープラーニングと人工知能の難問( 深化する機械学習-技術の進展とその応用)

Grid Pathfinding on the 2<i>k</i> Neighborhoods

Grid pathfinding, an old AI problem, is central for the development of navigation systems for autonomous agents. A surprising fact about the vast literature on this problem is that very limited neighborhoods have been studied. Indeed, only the 4- and 8-neighborhoods are usually considered, and rarely the 16-neighborhood. This paper describes three contributions that enable the construction of effective grid path planners for extended 2k-neighborhoods. First, we provide a simple recursive definition of the 2k-neighborhood in terms of the 2k–1-neighborhood. Second, we derive distance functions, for any k >1, which allow us to propose admissible heurisitics which are perfect for obstacle-free grids. Third, we describe a canonical ordering which allows us to implement a version of A* whose performance scales well when increasing k. Our empirical evaluation shows that the heuristics we propose are superior to the Euclidean distance (ED) when regular A* is used. For grids beyond 64 the overhead of computing the heuristic yields decreased time performance compared to the ED. We found also that a configuration of our A*-based implementation, without canonical orders, is competitive with the "any-angle" path planner Theta$^*$ both in terms of solution quality and runtime.

Efficient Parameter Importance Analysis via Ablation with Surrogates

To achieve peak performance, it is often necessary to adjust the parameters of a given algorithm to the class of problem instances to be solved; this is known to be the case for popular solvers for a broad range of AI problems, including AI planning, propositional satisfiability (SAT) and answer set programming (ASP). To avoid tedious and often highly sub-optimal manual tuning of such parameters by means of ad-hoc methods, general-purpose algorithm configuration procedures can be used to automatically find performance-optimizing parameter settings. While impressive performance gains are often achieved in this manner, additional, potentially costly parameter importance analysis is required to gain insights into what parameter changes are most responsible for those improvements. Here, we show how the running time cost of ablation analysis, a well-known general-purpose approach for assessing parameter importance, can be reduced substantially by using regression models of algorithm performance constructed from data collected during the configuration process. In our experiments, we demonstrate speed-up factors between 33 and 14 727 for ablation analysis on various configuration scenarios from AI planning, SAT, ASP and mixed integer programming (MIP).

Static and Dynamic Portfolio Methods for Optimal Planning: An Empirical Analysis

Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range of AI problems. Notably, portfolio techniques have been prominently applied to suboptimal (satisficing) AI planning. Here, we consider the construction of sequential planner portfolios for domainindependent optimal planning. Specifically, we introduce four techniques (three of which are dynamic) for per-instance planner schedule generation using problem instance features, and investigate the usefulness of a range of static and dynamic techniques for combining planners. Our extensive empirical analysis demonstrates the benefits of using static and dynamic sequential portfolios for optimal planning, and provides insights on the most suitable conditions for their fruitful exploitation.

Generation of Sokoban Stages using Recurrent Neural Networks

Puzzles and board games represent several important classes of AI problems, but also represent difficult complexity classes. In this paper, we propose a deep learning based alternative to train a neural network model to find solution states of the popular puzzle game Sokoban. The network trains against a classical solver that uses theorem proving as the oracle of valid and invalid games states, in a setup that is similar to the popular adversarial training framework. Using our approach, we have been able to verify the validity of a Sokoban puzzle up to an accuracy of 99% on the test set. We have also been able to train our network to generate the next possible state of the puzzle board up to an accuracy of 99% on the validation set. We hope that through this approach, a trained neural network will be able to replace human experts and classical rule-based AI in generating new instances and solutions for such games.