During the last decade multi-armed bandits (MAB) have received a great deal of attention in machine learning and related fields, due to their wide practical and theoretical importance.

England's 1,000-year-old legal system -- still steeped in traditions that include wearing wigs and robes -- has taken a cautious step into the future by giving judges permission to use artificial intelligence to help produce rulings. The Courts and Tribunals Judiciary last month said AI could help write opinions but stressed it shouldn't be used for research or legal analyses because the technology can fabricate information and provide misleading, inaccurate and biased information. "Judges do not need to shun the careful use of AI," said Master of the Rolls Geoffrey Vos, the second-highest ranking judge in England and Wales. "But they must ensure that they protect confidence and take full personal responsibility for everything they produce." At a time when scholars and legal experts are pondering a future when AI could replace lawyers, help select jurors or even decide cases, the approach spelled out Dec. 11 by the judiciary is restrained. But for a profession slow to embrace technological change, it's a proactive step as government and industry -- and society in general -- react to a rapidly advancing technology alternately portrayed as a panacea and a menace.

AI-enabled military systems have been effective in battle, but some reliability issues still concern troops and their commanders: former Air Force test pilot. Cheap drones equipped with AI can destroy expensive military equipment, and the Pentagon will need to incorporate autonomous technology into its strategy to advance into the next generation of warfare, a former test pilot and military tech company executive told Fox News. "What we've seen in Europe and other theaters is that they've democratized warfare," said EpiSci Vice President of Tactical Autonomous Systems Chris Gentile. "A $1,000 drone can take out a multimillion-dollar asset." The Pentagon has a portfolio of over 800 contracts for AI-enabled projects.

We naturally generalize the on-line graph prediction problem to a version of stochastic contextual bandit problems where contexts are vertices in a graph and the structure of the graph provides information on the similarity of contexts. More specifically, we are given a graph $G=(V,E)$, whose vertex set $V$ represents contexts with {\em unknown} vertex label $y$. In our stochastic contextual bandit setting, vertices with the same label share the same reward distribution. The standard notion of instance difficulties in graph label prediction is the cutsize $f$ defined to be the number of edges whose end points having different labels. For line graphs and trees we present an algorithm with regret bound of $\tilde{O}(T^{2/3}K^{1/3}f^{1/3})$ where $K$ is the number of arms. Our algorithm relies on the optimal stochastic bandit algorithm by Zimmert and Seldin~[AISTAT'19, JMLR'21]. When the best arm outperforms the other arms, the regret improves to $\tilde{O}(\sqrt{KT\cdot f})$. The regret bound in the later case is comparable to other optimal contextual bandit results in more general cases, but our algorithm is easy to analyze, runs very efficiently, and does not require an i.i.d. assumption on the input context sequence. The algorithm also works with general graphs using a standard random spanning tree reduction.

Information Extraction (IE) is the technique for transforming unstructured textual data into structured representation that can be understood by machines. The exponential growth of the Web generates an exceptional quantity of data for which automatic knowledge capture is essential. This work describes the methodology for Web scale Information Extraction adopted by the LODIE project (Linked Open Data Information Extraction). LODIE aims to develop Information Extraction techniques able to (i) scale at web level and (ii) adapt to user information need. The core idea behind LODIE is the usage of Linked Open Data, a very large-scale information resource, as a ground-breaking solution for IE, which provides invaluable annotated data on a growing number of domains.

This content can also be viewed on the site it originates from. On a cloudless morning last May, a pilot took off from the Niagara Falls International Airport, heading for restricted military airspace over Lake Ontario. The plane, which bore the insignia of the United States Air Force, was a repurposed Czechoslovak jet, an L-39 Albatros, purchased by a private defense contractor. The bay in front of the cockpit was filled with sensors and computer processors that recorded the aircraft's performance. For two hours, the pilot flew counterclockwise around the lake.

We study the multi-armed bandit (MAB) problem with composite and anonymous feedback. In this model, the reward of pulling an arm spreads over a period of time (we call this period as reward interval) and the player receives partial rewards of the action, convoluted with rewards from pulling other arms, successively. Existing results on this model require prior knowledge about the reward interval size as an input to their algorithms. In this paper, we propose adaptive algorithms for both the stochastic and the adversarial cases, without requiring any prior information about the reward interval. For the stochastic case, we prove that our algorithm guarantees a regret that matches the lower bounds (in order). For the adversarial case, we propose the first algorithm to jointly handle non-oblivious adversary and unknown reward interval size. We also conduct simulations based on real-world dataset. The results show that our algorithms outperform existing benchmarks.

In the Report, “[Meta-analysis reveals declines in terrestrial but increases in freshwater insect abundances][1],” the following corrections have been made. After publication, some errors in the data underlying the analyses, and the processing of it, were brought to the authors’ attention. The most important was a mistake in the processing of the Environmental Change Network moth data ([ 1 ][2]) from the UK (Datasource_ID 1006). The authors made two errors in processing these data: (i) They neglected to recode abundance counts with error code ‘101’ (indicating no sample taken) as missing data values and they instead entered into the analysis as values of 101, and (ii) they did not sufficiently account for the change in sampling protocol over time; moths were sampled all nights of the year in the first years but only on nights with favorable weather in later years. This led to higher average counts per night in later years because there were fewer data from the nights with low moth counts. These two errors produced a false-positive trend for this dataset. The authors have fixed these issues by removing the error code and retaining only the summer months during which sampling was consistent over time. Furthermore, they revisited all the other datasets in the analysis to check for any other errors in the data, including error codes, missing zeros, duplicate values, outliers, and sampling effort consistency across plots, and corrected these when necessary. They found inconsistencies in the source data of dataset 502 ([ 2 ][3]) and removed all years with missing species. As a result, the authors excluded 8 (out of 30) plots from this dataset because they no longer met the inclusion criteria. Dataset 1424 ([ 3 ][4]) was duplicated in dataset 1347 ([ 4 ][5]) and was thus removed because the latter provided more years of data. The authors retained 165 datasets and 1668 plots. In all, they made changes to 22 of these datasets. All corrections and their effect on the random-effects estimate of each dataset are detailed in the supplementary materials, and all figures and tables in the supplementary materials, as well as in data S1 and S2 and in the repository ([ 5 ][6]), have been replaced. It was also brought to the authors’ attention that they should have been clearer regarding exclusion of non-insects from datasets comprising both insects and non-insect invertebrates, as well as datasets with variable sampling frequencies. They have now added an additional explanation to the methods section of the supplementary materials. In brief, they excluded non-insect invertebrate data as much as possible but not at the cost of also excluding insects. The authors have rerun all models presented in the original paper with the corrected data and found that none of the major qualitative conclusions of the paper changed. The quantitative estimates have changed somewhat, however: The average decline for terrestrial insects across all data are now –1.11% per year (–10.56% per decade) and the increase for freshwater insects is now +1.16% per year (+12.24% per decade), both well within the 95% credible intervals of the previous estimates. In the geographic analysis, Europe now shows weak evidence for a decline of terrestrial insects of –0.76% per year (–7.3% per decade, P = 0.947), which is perpetuated across all time slices of Fig. 3 in the paper (ranging between moderate and strong evidence). Overall, the authors found more strengthening of trends than weakening of trends. For example, there is now weak evidence for a decline of terrestrial biomass and for a positive effect of increasing temperatures on terrestrial insect abundances. They also found weak evidence for a negative effect of last year of sampling on the trend estimates, suggesting that trends are more negative in datasets with more recent data. This matches the progressively more negative trends in the European terrestrial data. All old and new model estimates, presented as the percentage change per year, and a detailed description of the changes to the materials and methods, can be found on Zenodo (). 1. [↵][7]S. Rennie, J. Adamson, R. Anderson, C. Andrews, J. Bater, N. Bayfield, K. Beaton, D. Beaumont, S. Benham, V. Bowmaker, C. Britt, R. Brooker, D. Brooks, J. Brunt, G. Common, R. Cooper, S. Corbett, N. Critchley, P. Dennis, J. Dick, B. Dodd, N. Dodd, N. Donovan, J. Easter, M. Flexen, A. Gardiner, D. Hamilton, P. Hargreaves, M. Hatton-Ellis, M. Howe, J. Kahl, M. Lane, S. Langan, D. Lloyd, B. McCarney, Y. McElarney, C. McKenna, S. McMillan, F. Milne, L. Milne, M. Morecroft, M. Murphy, A. Nelson, H. Nicholson, D. Pallett, D. Parry, I. Pearce, G. Pozsgai, A. Riley, R. Rose, S. Schafer, T. Scott, L. Sherrin, C. Shortall, R. Smith, P. Smith, R. Tait, C. Taylor, M. Taylor, M. Thurlow, A. Turner, K. Tyson, H. Watson, M. Whittaker, I. Woiwod, C. Wood, UK Environmental Change Network (ECN) Moth Data: 1992-2015, NERC Environmental Information Data Centre (2018); . 2. [↵][8]NERC Centre for Population Biology Imperial College, Global population dynamics database, Version 2 (2010); . 3. [↵][9]1. J. M. McCarthy, 2. C. L. Hein, 3. J. D. Olden, 4. M. J. Vander Zanden , Coupling long-term studies with meta-analysis to investigate impacts of non-native crayfish on zoobenthic communities. Freshw. Biol. 51, 224–235 (2006). 10.1111/j.1365-2427.2005.01485.x [OpenUrl][10][CrossRef][11] 4. [↵][12]J. Magnuson, C. S. E. Stanley, North Temperate Lakes LTER: Benthic macroinvertebrates 1981-current, Environmental Data Initiative (2010); . 5. [↵][13]R. van Klink, D. E. Bowler, J. M. Chase, O. Comay, M. M. Driessen, S. K. M. Ernest, A. Gentile, F. Gilbert, K. B. Gongalky, G. Pe’er, I. Pe’er, V. H. Resh, A. B. Swengel, S. R. Swengel, T. J. Valone, R. Vermeulen, T. Wepprich, J. Wiedmann, A global database of long-term changes in insect assemblages, Knowledge Network for Biocomplexity (KNB) (2020); . [1]: https://science.sciencemag.org/content/368/6489/417 [2]: #ref-1 [3]: #ref-2 [4]: #ref-3 [5]: #ref-4 [6]: #ref-5 [7]: #xref-ref-1-1 "View reference 1 in text" [8]: #xref-ref-2-1 "View reference 2 in text" [9]: #xref-ref-3-1 "View reference 3 in text" [10]: {openurl}?query=rft.jtitle%253DFreshw.%2BBiol.%26rft.volume%253D51%26rft.spage%253D224%26rft_id%253Dinfo%253Adoi%252F10.1111%252Fj.1365-2427.2005.01485.x%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [11]: /lookup/external-ref?access_num=10.1111/j.1365-2427.2005.01485.x&link_type=DOI [12]: #xref-ref-4-1 "View reference 4 in text" [13]: #xref-ref-5-1 "View reference 5 in text"