DeepBrain: Functional Representation of Neural In-Situ Hybridization Images for Gene Ontology Classification Using Deep Convolutional Autoencoders

This paper presents a novel deep learning-based method for learning afunctional representation of mammalian neural images. The method uses a deepconvolutional denoising autoencoder (CDAE) for generating an invariant, compactrepresentation of in situ hybridization (ISH) images. While most existingmethods for bio-imaging analysis were not developed to handle images withhighly complex anatomical structures, the results presented in this paper showthat functional representation extracted by CDAE can help learn features offunctional gene ontology categories for their classification in a highlyaccurate manner. Using this CDAE representation, our method outperforms theprevious state-of-the-art classification rate, by improving the average AUCfrom 0.92 to 0.98, i.e., achieving 75% reduction in error. The method operateson input images that were downsampled significantly with respect to theoriginal ones to make it computationally feasible.

Expert-Driven Genetic Algorithms for Simulating Evaluation Functions

In this paper we demonstrate how genetic algorithms can be used to reverseengineer an evaluation function's parameters for computer chess. Our resultsshow that using an appropriate expert (or mentor), we can evolve a program thatis on par with top tournament-playing chess programs, outperforming a two-timeWorld Computer Chess Champion. This performance gain is achieved by evolving aprogram that mimics the behavior of a superior expert. The resulting evaluationfunction of the evolved program consists of a much smaller number of parametersthan the expert's. The extended experimental results provided in this paperinclude a report of our successful participation in the 2008 World ComputerChess Championship. In principle, our expert-driven approach could be used in awide range of problems for which appropriate experts are available.

An Automatic Solver for Very Large Jigsaw Puzzles Using Genetic Algorithms

In this paper we propose the first effective genetic algorithm (GA)-basedjigsaw puzzle solver. We introduce a novel crossover procedure that merges two"parent" solutions to an improved "child" configuration by detecting,extracting, and combining correctly assembled puzzle segments. The solverproposed exhibits state-of-the-art performance, as far as handling previouslyattempted puzzles more accurately and efficiently, as well puzzle sizes thathave not been attempted before. The extended experimental results provided inthis paper include, among others, a thorough inspection of up to 30,745-piecepuzzles (compared to previous attempts on 22,755-piece puzzles), using aconsiderably faster concurrent implementation of the algorithm. Furthermore, weexplore the impact of different phases of the novel crossover operator byexperimenting with several variants of the GA. Finally, we compare differentfitness functions and their effect on the overall results of the GA-basedsolver.

The West Bank and International Humanitarian Law on the Eve of the Fiftieth Anniversary of the Six-Day War

The West Bank and the Settlements, again? Readers may have had enough of this subject. But these are exceptional times. The adoption by the Security Council of Resolution 2334 on December 23, 2016, the unprecedented speech by Secretary Kerry delivered shortly thereafter, and the immediate rejection of both by Prime Minister Netanyahu, combined with the approach of the fiftieth anniversary of the Six-Day War in June 2017 and the continued march toward an inexorable demographic change in the West Bank, not to mention the nomination as U.S. Ambassador to Israel of a person reportedly supporting an active settlement policy and annexation: the confluence of these events demands our renewed attention. And while these developments undoubtedly have powerful political dimensions, they also call upon those of us who care about international law to speak up in support of its requirements and application.

Introduction to Symposium on Revisiting Israel's Settlements

This symposium accompanies Theodor Meron’s editorial comment, to be published in the April issue of AJIL, addressing the legal status of the West Bank and the applicable rules of international humanitarian law, particularly the Fourth Geneva Convention’s prohibition on settlements on occupied territories in Article 49(6) and the Hague Convention (IV) Respecting the Laws and Customs of War on Land.1 In that editorial comment, which is being published alongside this symposium on “First View,”Meron revisits opinions he first expressed, as a Legal Adviser to the Israel Ministry of Foreign Affairs, immediately after the Six-Day War. Those legal memoranda have been recently unearthed and published.2 Like Meron, the AJIL Unbound board believes that the fiftieth anniversary of that conflict and other recent events—including the adoption by the Security Council of Resolution 2334 on December 23, 2016 followed by a controversial speech delivered by out-going Secretary of State John Kerry that provoked outrage by then President-elect Donald Trump, Prime Minister Benjamin Netanyahu, some members of Congress, and others3—merits revisiting some legal questions regarding Israeli settlements. These evergreen issues, never far from view, are unquestionably again front and center. They are clearly so for a new U.S. President inclined to disrupt seemingly settled verities on the Middle East—and even for contemporary theatre audiences thanks to a critically acclaimed new play, Oslo, by J.T. Rogers that reexamines the secret behind-thescenes negotiations that ultimately led to the Oslo Accords of 1993. This symposium’s six authors address a number of issues to which Meron only alludes. Eyal Benvenisti examines the 1968 legal article that Meron mentions, namely Yehuda Blum’s “The Missing Reversioner: Reflections on the Status of Judea and Samaria”; he argues that its subsequent use by Israel’s thenMilitary Advocate General Meir Shamgar fundamentally altered Israel’s initial settlement policies and changed the course of history.4 Pnina Sharvit Baruch tacklesMeron’s core contention that the laws of occupation have a predominately “humanitarian” (or what he calls a “people-oriented”) purpose. In her view, those rules also have a “sovereign dimension” that cannot be ignored insofar as they seek to preserve pre-existing claims or rights of the occupier as well as those of the original sovereign of disputed territory.5 David Kretzmer draws on his recent book on the subject to examine how the Israeli Supreme Court has broached the legality and consequences of settlements, particularly for Israeli citizens themselves, and also to compare that Court’s approach with that of the International Court of Justice.6 The

A “Supertanker” Against Bureaucracy in the Wake of a Housing Crisis: Neoliberalizing Planning in Netanyahu's Israel

This paper critically questions the state's hostile takeover of planning regulation followed by experimentation initiated by the Israeli Prime Minister Benjamin Netanyahu who has been seeking to subordinate the planning apparatus to market calculus and to short‐term political ends. To substantiate this argument, I have examined a large corpus of documents (official government documents, planning records, and court appeals and rulings, and NGO reports) and analyzed the media coverage between 2011 and 2016. By introducing fast‐track planning that is firmly controlled by the central state and focusing on the fictitious delivery of housing units, the structure of the planning regulation has dramatically changed. Further, two already‐dominant government ministries (Finance and Defense) have been significantly empowered, becoming the supervisors of the reformed planning system. In a state captivated by neoliberal fixation and embroiled in a housing crisis, the restructuring of planning governance has been a means to an end.

UN‐ Al Sisi Meets Netanyahu

Genetic Algorithms for Mentor-Assisted Evaluation Function Optimization

In this paper we demonstrate how genetic algorithms can be used to reverseengineer an evaluation function's parameters for computer chess. Our resultsshow that using an appropriate mentor, we can evolve a program that is on parwith top tournament-playing chess programs, outperforming a two-time WorldComputer Chess Champion. This performance gain is achieved by evolving aprogram with a smaller number of parameters in its evaluation function to mimicthe behavior of a superior mentor which uses a more extensive evaluationfunction. In principle, our mentor-assisted approach could be used in a widerange of problems for which appropriate mentors are available.

Simulating Human Grandmasters: Evolution and Coevolution of Evaluation Functions

This paper demonstrates the use of genetic algorithms for evolving agrandmaster-level evaluation function for a chess program. This is achieved bycombining supervised and unsupervised learning. In the supervised learningphase the organisms are evolved to mimic the behavior of human grandmasters,and in the unsupervised learning phase these evolved organisms are furtherimproved upon by means of coevolution. While past attempts succeeded in creating a grandmaster-level program bymimicking the behavior of existing computer chess programs, this paper presentsthe first successful attempt at evolving a state-of-the-art evaluation functionby learning only from databases of games played by humans. Our resultsdemonstrate that the evolved program outperforms a two-time World ComputerChess Champion.

Genetic Algorithm-Based Solver for Very Large Multiple Jigsaw Puzzles of Unknown Dimensions and Piece Orientation

In this paper we propose the first genetic algorithm (GA)-based solver forjigsaw puzzles of unknown puzzle dimensions and unknown piece location andorientation. Our solver uses a novel crossover technique, and sets a newstate-of-the-art in terms of the puzzle sizes solved and the accuracy obtained.The results are significantly improved, even when compared to previous solversassuming known puzzle dimensions. Moreover, the solver successfully contendswith a mixed bag of multiple puzzle pieces, assembling simultaneously allpuzzles.