Risk is 6 player game with significant randomness and a large game-tree complexity which poses a challenge to creating an agent to play the game effectively. Previous AIs focus on creating high-level handcrafted features determine agent decision making. In this project, I create D.A.D, A Risk agent using temporal difference reinforcement learning to train a Deep Neural Network including a Graph Convolutional Network to evaluate player positions. This is used in a game-tree to select optimal moves. This allows minimal handcrafting of knowledge into the AI, assuring input features are as low-level as possible to allow the network to extract useful and sophisticated features itself, even with the network starting from a random initialisation. I also tackle the issue of non-determinism in Risk by introducing a new method of interpreting attack moves necessary for the search. The result is an AI which wins 35% of the time versus 5 of best inbuilt AIs in Lux Delux, a Risk variant.

In this article, to answer these questions, we go through the Monte Carlo Tree Search fundamentals. Since in the next articles, we will implement this algorithm on "HEX" board game, I try to explain the concepts through examples in this board game environment. If you're more interested in the code, find it in this link. There is also a more optimized version which is applicable on linux due to utilizing cython and you can find it in here. Monte Carlo method was coined by Stanislaw Ulam for the first time after applying statistical approach "The Monte Carlo method".

We study the behavior of several black-box search algorithms used for generating adversarial examples for natural language processing (NLP) tasks. We perform a fine-grained analysis of three elements relevant to search: search algorithm, search space, and search budget. When new search methods are proposed in past work, the attack search space is often modified alongside the search method. Without ablation studies benchmarking the search algorithm change with the search space held constant, an increase in attack success rate could from an improved search method or a less restrictive search space. Additionally, many previous studies fail to properly consider the search algorithms' run-time cost, which is essential for downstream tasks like adversarial training. Our experiments provide a reproducible benchmark of search algorithms across a variety of search spaces and query budgets to guide future research in adversarial NLP. Based on our experiments, we recommend greedy attacks with word importance ranking when under a time constraint or attacking long inputs, and either beam search or particle swarm optimization otherwise. Code implementation shared via https://github.com/QData/TextAttack

Limited power and computational resources, absence of high-end sensor equipment and GPS-denied environments are challenges faced by autonomous micro areal vehicles (MAVs). We address these challenges in the context of autonomous navigation and landing of MAVs in indoor environments and propose a vision-based control approach using Supervised Learning. To achieve this, we collected data samples in a simulation environment which were labelled according to the optimal control command determined by a path planning algorithm. Based on these data samples, we trained a Convolutional Neural Network (CNN) that maps low resolution image and sensor input to high-level control commands. We have observed promising results in both obstructed and non-obstructed simulation environments, showing that our model is capable of successfully navigating a MAV towards a landing platform. Our approach requires shorter training times than similar Reinforcement Learning approaches and can potentially overcome the limitations of manual data collection faced by comparable Supervised Learning approaches.

Recent developments in Omics-technologies revolutionized the investigation of biology by producing molecular data in multiple dimensions and scale. This breakthrough in biology raises the crucial issue of their interpretation based on modelling. In this undertaking, network provides a suitable framework for modelling the interactions between molecules. Basically a Biological network is composed of nodes referring to the components such as genes or proteins, and the edges/arcs formalizing interactions between them. The evolution of the interactions is then modelled by the definition of a dynamical system. Among the different categories of network, the Boolean network offers a reliable qualitative framework for the modelling. Automatically synthesizing a Boolean network from experimental data therefore remains a necessary but challenging issue. In this study, we present taboon, an original work-flow for synthesizing Boolean Networks from biological data. The methodology uses the data in the form of Boolean profiles for inferring all the potential local formula inference. They combine to form the model space from which the most truthful model with regards to biological knowledge and experiments must be found. In the taboon work-flow the selection of the fittest model is achieved by a Tabu-search algorithm. taboon is an automated method for Boolean Network inference from experimental data that can also assist to evaluate and optimize the dynamic behaviour of the biological networks providing a reliable platform for further modelling and predictions.

We investigate the autonomous navigation of a mobile robot in the presence of other moving vehicles under time-varying uncertain environmental disturbances. We first predict the future state distributions of other vehicles to account for their uncertain behaviors affected by the time-varying disturbances. We then construct a dynamic-obstacle-aware reachable space that contains states with high probabilities to be reached by the robot, within which the optimal policy is searched. Since, in general, the dynamics of both the vehicle and the environmental disturbances are nonlinear, we utilize a nonlinear Gaussian filter -- the unscented transform -- to approximate the future state distributions. Finally, the forward reachable space computation and backward policy search are iterated until convergence. Extensive simulation evaluations have revealed significant advantages of this proposed method in terms of computation time, decision accuracy, and planning reliability.

Under high complexity - given by pervasive interdependence between constituent elements of a decision in an NK landscape - our algorithm obtains fitness superior to that reported in extant research. We distribute the decision elements comprising a decision into clusters. When a change in value of a decision element is considered, a forward move is made if the aggregate fitness of the cluster members residing alongside the decision element is higher. The decision configuration with the highest fitness in the path is selected. Increasing the number of clusters obtains even higher fitness. Further, implementing moves comprising of up to two changes in a cluster also obtains higher fitness. Our algorithm obtains superior outcomes by enabling more extensive search, allowing inspection of more distant configurations. We name this algorithm the muddling through algorithm, in memory of Charles Lindblom who spotted the efficacy of the process long before sophisticated computer simulations came into being.

Combinatorial optimization plays an important role in real-world problem solving. In the big data era, the dimensionality of a combinatorial optimization problem is usually very large, which poses a significant challenge to existing solution methods. In this paper, we examine the generalization capability of a machine learning model for problem reduction on the classic travelling salesman problems (TSP). We demonstrate that our method can greedily remove decision variables from an optimization problem that are predicted not to be part of an optimal solution. More specifically, we investigate our model's capability to generalize on test instances that have not been seen during the training phase. We consider three scenarios where training and test instances are different in terms of: 1) problem characteristics; 2) problem sizes; and 3) problem types. Our experiments show that this machine learning based technique can generalize reasonably well over a wide range of TSP test instances with different characteristics or sizes. While the accuracy of predicting unused variables naturally deteriorates as a test instance is further away from the training set, we observe that even when tested on a different TSP problem variant, the machine learning model still makes useful predictions about which variables can be eliminated without significantly impacting solution quality.

The widespread availability of superhuman AI engines is changing how we play the ancient game of Go. The open-source software packages developed after the AlphaGo series shifted focus from producing strong playing entities to providing tools for analyzing games. Here we describe two ways of how the innovations of the second generation engines (e.g.~score estimates, variable komi) can be used for defining new metrics that help deepen our understanding of the game. First, we study how much information the search component contributes in addition to the raw neural network policy output. This gives an intrinsic strength measurement for the neural network. Second, we define the effect of a move by the difference in score estimates. This gives a fine-grained, move-by-move performance evaluation of a player. We use this in combating the new challenge of detecting online cheating.

We propose a novel method for expediting both symmetric and asymmetric Distributed Constraint Optimization Problem (DCOP) solvers. The core idea is based on initializing DCOP solvers with greedy fast non-iterative DCOP solvers. This is contrary to existing methods where initialization is always achieved using a random value assignment. We empirically show that changing the starting conditions of existing DCOP solvers not only reduces the algorithm convergence time by up to 50\%, but also reduces the communication overhead and leads to a better solution quality. We show that this effect is due to structural improvements in the variable assignment, which is caused by the spreading pattern of DCOP algorithm activation.) /Subject (Hybrid DCOPs)