File size: 139,862 Bytes
6fa4bc9 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 | {
"paper_id": "P16-1013",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T09:00:23.461571Z"
},
"title": "Learning the Curriculum with Bayesian Optimization for Task-Specific Word Representation Learning",
"authors": [
{
"first": "Yulia",
"middle": [],
"last": "Tsvetkov",
"suffix": "",
"affiliation": {},
"email": "ytsvetko@cs.cmu.edu"
},
{
"first": "Manaal",
"middle": [],
"last": "Faruqui",
"suffix": "",
"affiliation": {},
"email": "mfaruqui@cs.cmu.edu"
},
{
"first": "\u2660",
"middle": [],
"last": "Wang",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Brian",
"middle": [],
"last": "Macwhinney",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Chris",
"middle": [],
"last": "Dyer",
"suffix": "",
"affiliation": {},
"email": "cdyer@cs.cmu.edu"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "We use Bayesian optimization to learn curricula for word representation learning, optimizing performance on downstream tasks that depend on the learned representations as features. The curricula are modeled by a linear ranking function which is the scalar product of a learned weight vector and an engineered feature vector that characterizes the different aspects of the complexity of each instance in the training corpus. We show that learning the curriculum improves performance on a variety of downstream tasks over random orders and in comparison to the natural corpus order.",
"pdf_parse": {
"paper_id": "P16-1013",
"_pdf_hash": "",
"abstract": [
{
"text": "We use Bayesian optimization to learn curricula for word representation learning, optimizing performance on downstream tasks that depend on the learned representations as features. The curricula are modeled by a linear ranking function which is the scalar product of a learned weight vector and an engineered feature vector that characterizes the different aspects of the complexity of each instance in the training corpus. We show that learning the curriculum improves performance on a variety of downstream tasks over random orders and in comparison to the natural corpus order.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "It is well established that in language acquisition, there are robust patterns in the order by which phenomena are acquired. For example, prototypical concepts are acquired earlier; concrete words tend to be learned before abstract ones (Rosch, 1978) . The acquisition of lexical knowledge in artificial systems proceeds differently. In general, models will improve during the course of parameter learning, but the time course of acquisition is not generally studied beyond generalization error as a function of training time or data size. We revisit this issue of choosing the order of learning-curriculum learning-framing it as an optimization problem so that a rich array of factors-including nuanced measures of difficulty, as well as prototypicality and diversity-can be exploited.",
"cite_spans": [
{
"start": 237,
"end": 250,
"text": "(Rosch, 1978)",
"ref_id": "BIBREF33"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Prior research focusing on curriculum strategies in NLP is scarce, and has conventionally been following a paradigm of \"starting small\" (Elman, 1993) , i.e., initializing the learner with \"simple\" examples first, and then gradually increasing data complexity (Bengio et al., 2009; Spitkovsky et al., 2010) . In language modeling, this preference for increasing complexity has been realized by curricula that increase the entropy of training data by growing the size of the training vocabulary from frequent to less frequent words (Bengio et al., 2009) . In unsupervised grammar induction, an effective curriculum comes from increasing length of training sentences as training progresses (Spitkovsky et al., 2010) . These case studies have demonstrated that carefully designed curricula can lead to better results. However, they have relied on heuristics in selecting curricula or have followed the intuitions of human and animal learning (Kail, 1990; Skinner, 1938) . Had different heuristics been chosen, the results would have been different. In this paper, we use curriculum learning to create improved word representations. However, rather than testing a small number of curricula, we search for an optimal curriculum using Bayesian optimization. A curriculum is defined to be the ordering of the training instances, in our case it is the ordering of paragraphs in which the representation learning model reads the corpus. We use a linear ranking function to conduct a systematic exploration of interacting factors that affect curricula of representation learning models. We then analyze our findings, and compare them to human intuitions and learning principles.",
"cite_spans": [
{
"start": 136,
"end": 149,
"text": "(Elman, 1993)",
"ref_id": "BIBREF8"
},
{
"start": 259,
"end": 280,
"text": "(Bengio et al., 2009;",
"ref_id": "BIBREF1"
},
{
"start": 281,
"end": 305,
"text": "Spitkovsky et al., 2010)",
"ref_id": "BIBREF42"
},
{
"start": 530,
"end": 551,
"text": "(Bengio et al., 2009)",
"ref_id": "BIBREF1"
},
{
"start": 687,
"end": 712,
"text": "(Spitkovsky et al., 2010)",
"ref_id": "BIBREF42"
},
{
"start": 938,
"end": 950,
"text": "(Kail, 1990;",
"ref_id": "BIBREF18"
},
{
"start": 951,
"end": 965,
"text": "Skinner, 1938)",
"ref_id": "BIBREF39"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "We treat curriculum learning as an outer loop in the process of learning and evaluation of vectorspace representations of words; the iterative procedure is (1) predict a curriculum; (2) train word embeddings; (3) evaluate the embeddings on tasks that use word embeddings as the sole features. Through this model we analyze the impact of curriculum on word representation models and on extrinsic tasks. To quantify curriculum properties, we define three groups of features aimed at analyzing statistical and linguistic content and structure of training data: (1) diversity, (2) simplicity, and (3) prototypicality. A function of these features is computed to score each paragraph in the training data, and the curriculum is determined by sorting corpus paragraphs by the paragraph scores. We detail the model in \u00a72. Word vectors are learned from the sorted corpus, and then evaluated on partof-speech tagging, parsing, named entity recognition, and sentiment analysis ( \u00a73). Our experiments confirm that training data curriculum affects model performance, and that models with optimized curriculum consistently outperform baselines trained on shuffled corpora ( \u00a74). We analyze our findings in \u00a75.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The contributions of this work are twofold. First, this is the first framework that formulates curriculum learning as an optimization problem, rather then shuffling data or relying on human intuitions. We experiment with optimizing the curriculum of word embeddings, but in principle the curriculum of other models can be optimized in a similar way. Second, to the best of our knowledge, this study is the first to analyze the impact of distributional and linguistic properties of training texts on the quality of task-specific word embeddings.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "We are considering the problem of maximizing a performance of an NLP task through sequentially optimizing the curriculum of training data of word vector representations that are used as features in the task.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Curriculum Learning Model",
"sec_num": "2"
},
{
"text": "Let X = {x 1 , x 2 , . . . , x n } be the training corpus with n lines (sentences or paragraphs). The curriculum of word representations is quantified by scoring each of the paragraphs according to the linear function w \u03c6(X ), where \u03c6(X ) \u2208 R \u00d71 is a real-valued vector containing linguistic features extracted for each paragraph, and w \u2208 R \u00d71 denote the weights learned for these features. The feature values \u03c6(X ) are z-normalized across all paragraphs. These scores are used to specify the order of the paragraphs in the corpus-the curriculum: we sort the paragraphs by their scores.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Curriculum Learning Model",
"sec_num": "2"
},
{
"text": "After the paragraphs are curriculum-ordered, the reordered training corpus is used to generate word representations. These word representations are then used as features in a subsequent NLP task. We define the objective function eval : X \u2192 R, which is the quality estimation metric for this NLP task performed on a held-out dataset (e.g., corre-lation, accuracy, F 1 score, BLEU). Our goal is to define the features \u03c6(X ) and to find the optimal weights w that maximize eval.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Curriculum Learning Model",
"sec_num": "2"
},
{
"text": "We optimize the feature weights using Bayesian optimization; we detail the model in \u00a72.1. Distributional and linguistic features inspired by prior research in language acquisition and second language learning are described in \u00a72.2. Figure 1 shows the computation flow diagram. ",
"cite_spans": [],
"ref_spans": [
{
"start": 232,
"end": 240,
"text": "Figure 1",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "Curriculum Learning Model",
"sec_num": "2"
},
{
"text": "As no assumptions are made regarding the form of eval(w), gradient-based methods cannot be applied, and performing a grid search over parameterizations of w would require a exponentially growing number of parameterizations to be traversed. Thus, we propose to use Bayesian Optimization (BayesOpt) as the means to maximize eval(w). BayesOpt is a methodology to globally optimize expensive, multimodal black-box functions (Shahriari et al., 2016; Bergstra et al., 2011; Snoek et al., 2012) . It can be viewed as a sequential approach to performing a regression from high-level model parameters (e.g., learning rate, number of layers in a neural network, and in our model-curriculum weights w) to the loss function or the performance measure (eval). An arbitrary objective function, eval, is treated as a black-box, and BayesOpt uses Bayesian inference to characterize a posterior distribution over functions that approximate eval. This model of eval is called the surrogate model. Then, the BayesOpt exploits this model to make decisions about eval, e.g., where is the expected maximum of the function, and what is the expected improvement that can be obtained over the best iteration so far. The strategy function, estimating the next set of parameters to explore given the current beliefs about eval is called the acquisition function. The surrogate model and the acquisition function are the two key components in the BayesOpt framework; their interaction is shown in Algorithm 1.",
"cite_spans": [
{
"start": 420,
"end": 444,
"text": "(Shahriari et al., 2016;",
"ref_id": "BIBREF37"
},
{
"start": 445,
"end": 467,
"text": "Bergstra et al., 2011;",
"ref_id": "BIBREF2"
},
{
"start": 468,
"end": 487,
"text": "Snoek et al., 2012)",
"ref_id": "BIBREF40"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Bayesian Optimization for Curriculum Learning",
"sec_num": "2.1"
},
{
"text": "The surrogate model allows us to cheaply approximate the quality of a set of parameters w without running eval(w), and the acquisition function uses this surrogate to choose a new value of w. However, a trade-off must be made: should the acquisition function move w into a region where the surrogate believes an optimal value will be found, or should it explore regions of the space that reveal more about how eval behaves, perhaps discovering even better values? That is, acquisition functions balance a tradeoff between exploration-by selecting w in the regions where the uncertainty of the surrogate model is high, and exploitation-by querying the regions where the model prediction is high.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Bayesian Optimization for Curriculum Learning",
"sec_num": "2.1"
},
{
"text": "Popular choices for the surrogate model are Gaussian Processes (Rasmussen, 2006; Snoek et al., 2012, GP) , providing convenient and powerful prior distribution on functions, and tree-structured Parzen estimators (Bergstra et al., 2011, TPE), tailored to handle conditional spaces. Choices of the acquisition functions include probability of improvement (Kushner, 1964) , expected improvement (EI) (Mo\u010dkus et al., 1978; Jones, 2001 ), GP upper confidence bound (Srinivas et al., 2010) , Thompson sampling (Thompson, 1933) , entropy search (Hennig and Schuler, 2012) , and dynamic combinations of the above functions (Hoffman et al., 2011); see Shahriari et al. (2016) for an extensive comparison. Yogatama et al. (2015) found that the combination of EI as the acquisition function and TPE as the surrogate model performed favorably in Bayesian optimization of text representations; we follow this choice in our model.",
"cite_spans": [
{
"start": 63,
"end": 80,
"text": "(Rasmussen, 2006;",
"ref_id": "BIBREF32"
},
{
"start": 81,
"end": 104,
"text": "Snoek et al., 2012, GP)",
"ref_id": null
},
{
"start": 353,
"end": 368,
"text": "(Kushner, 1964)",
"ref_id": "BIBREF21"
},
{
"start": 397,
"end": 418,
"text": "(Mo\u010dkus et al., 1978;",
"ref_id": "BIBREF28"
},
{
"start": 419,
"end": 430,
"text": "Jones, 2001",
"ref_id": "BIBREF17"
},
{
"start": 460,
"end": 483,
"text": "(Srinivas et al., 2010)",
"ref_id": "BIBREF43"
},
{
"start": 504,
"end": 520,
"text": "(Thompson, 1933)",
"ref_id": "BIBREF46"
},
{
"start": 538,
"end": 564,
"text": "(Hennig and Schuler, 2012)",
"ref_id": "BIBREF13"
},
{
"start": 643,
"end": 666,
"text": "Shahriari et al. (2016)",
"ref_id": "BIBREF37"
},
{
"start": 696,
"end": 718,
"text": "Yogatama et al. (2015)",
"ref_id": "BIBREF51"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Bayesian Optimization for Curriculum Learning",
"sec_num": "2.1"
},
{
"text": "To characterize and quantify a curriculum, we define three categories of features, focusing on various distributional, syntactic, and semantic aspects of training data. We now detail the feature categories along with motivations for feature selection.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "DIVERSITY. Diversity measures capture the distributions of types in data. Entropy is the bestknown measure of diversity in statistical research, but there are many others (Tang et al., 2006; Gimpel et al., 2013) . ",
"cite_spans": [
{
"start": 171,
"end": 190,
"text": "(Tang et al., 2006;",
"ref_id": "BIBREF45"
},
{
"start": 191,
"end": 211,
"text": "Gimpel et al., 2013)",
"ref_id": "BIBREF11"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "S 0 \u2190 T P E",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "Initialize surrogate model 4: for t \u2190 1 to T do 5:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "w t \u2190 argmax w A(w; S t\u22121 , H) Predict w t by optimizing acquisition function 6: eval(w t ) Evaluate w t on extrinsic task 7: H \u2190 H \u222a (w t , eval(w t )) Update obser- vation history 8:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "Estimate S t given H 9: end for 10: return H are used in many contrasting fields, from ecology and biology (Rosenzweig, 1995; Magurran, 2013) , to economics and social studies (Stirling, 2007) . Diversity has been shown effective in related research on curriculum learning in language modeling, vision, and multimedia analysis (Bengio et al., 2009; Jiang et al., 2014) .",
"cite_spans": [
{
"start": 107,
"end": 125,
"text": "(Rosenzweig, 1995;",
"ref_id": "BIBREF34"
},
{
"start": 126,
"end": 141,
"text": "Magurran, 2013)",
"ref_id": "BIBREF25"
},
{
"start": 176,
"end": 192,
"text": "(Stirling, 2007)",
"ref_id": "BIBREF44"
},
{
"start": 327,
"end": 348,
"text": "(Bengio et al., 2009;",
"ref_id": "BIBREF1"
},
{
"start": 349,
"end": 368,
"text": "Jiang et al., 2014)",
"ref_id": "BIBREF15"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "Let p i and p j correspond to empirical frequencies of word types t i and t j in the training data. Let d ij correspond to their semantic similarity, calculated as the cosine similarity between embeddings of t i and t j learned from the training data. We annotate each paragraph with the following diversity features:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Number of word types: #types",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Type-token ratio:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "#types #tokens \u2022 Entropy: \u2212 i p i ln(p i ) \u2022 Simpson's index (Simpson, 1949): i p i 2 \u2022 Quadratic entropy (Rao, 1982): 1 i,j d ij p i p j",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "SIMPLICITY. Spitkovsky et al. (2010) have validated the utility of syntactic simplicity in curriculum learning for unsupervised grammar induction by showing that training on sentences in order of increasing lengths outperformed other orderings. We explore the simplicity hypothesis, albeit without prior assumptions on specific ordering of data, and extend it to additional simplicity/complexity measures of training data. Our features are inspired by prior research in second language acquisition, text simplification, and readability assessment (Schwarm and Ostendorf, 2005; Heilman et al., 2007; Pitler and Nenkova, 2008; Vajjala and Meurers, 2012) . We use an off-the-shelf syntactic parser 2 (Zhang and Clark, 2011) to parse our training corpus. Then, the following features are used to measure phonological, lexical, and syntactic complexity of training paragraphs:",
"cite_spans": [
{
"start": 547,
"end": 576,
"text": "(Schwarm and Ostendorf, 2005;",
"ref_id": "BIBREF35"
},
{
"start": 577,
"end": 598,
"text": "Heilman et al., 2007;",
"ref_id": "BIBREF12"
},
{
"start": 599,
"end": 624,
"text": "Pitler and Nenkova, 2008;",
"ref_id": "BIBREF29"
},
{
"start": 625,
"end": 651,
"text": "Vajjala and Meurers, 2012)",
"ref_id": "BIBREF49"
},
{
"start": 697,
"end": 720,
"text": "(Zhang and Clark, 2011)",
"ref_id": "BIBREF52"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 PROTOTYPICALITY. This is a group of semantic features that use insights from cognitive linguistics and child language acquisition. The goal is to characterize the curriculum of representation learning in terms of the curriculum of human language learning. We resort to the Prototype theory (Rosch, 1978) , which posits that semantic categories include more central (or prototypical) as well as less prototypical words. For example, in the ANIMAL category, dog is more prototypical than sloth (because dog is more frequent); dog is more prototypical than canine (because dog is more concrete); and dog is more prototypical than bull terrier (because dog is less specific). According to the theory, more prototypical words are acquired earlier. We use lexical semantic databases to operationalize insights from the prototype theory in the following semantic features; the features are computed on token level and averaged over paragraphs:",
"cite_spans": [
{
"start": 292,
"end": 305,
"text": "(Rosch, 1978)",
"ref_id": "BIBREF33"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Age of acquisition (AoA) of words was extracted from the crowd-sourced database, containing over 50 thousand English words (Kuperman et al., 2012) . For example, the AoA of run is 4.47 (years), of flee is 8.33, and of abscond is 13.36. If a word was not found in the database it was assigned the maximal age of 25.",
"cite_spans": [
{
"start": 125,
"end": 148,
"text": "(Kuperman et al., 2012)",
"ref_id": "BIBREF20"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Concreteness ratings on the scale of 1-5 (1 is most abstract) for 40 thousand English lemmas (Brysbaert et al., 2014) . For example, cookie is rated as 5, and spirituality as 1.07.",
"cite_spans": [
{
"start": 95,
"end": 119,
"text": "(Brysbaert et al., 2014)",
"ref_id": "BIBREF3"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Imageability ratings are taken from the MRC psycholinguistic database (Wilson, 1988) . Following Tsvetkov et al. (2014) , we used the MRC annotations as seed, and propagated the ratings to all vocabulary words using the word embeddings as features in an 2 -regularized logistic regression classifier.",
"cite_spans": [
{
"start": 72,
"end": 86,
"text": "(Wilson, 1988)",
"ref_id": "BIBREF50"
},
{
"start": 99,
"end": 121,
"text": "Tsvetkov et al. (2014)",
"ref_id": "BIBREF48"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Conventionalization features count the number of \"conventional\" words and phrases in a paragraph. Assuming that a Wikipedia title is a proxy to a conventionalized concept, we counted the number of existing titles (from a database of over 4.5 million titles) in the paragraph.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Number of syllables scores are also extracted from the AoA database; out-of-database words were annotated as 5-syllable words.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Relative frequency in a supersense was computed by marginalizing the word frequencies in the training corpus over coarse semantic categories defined in the WordNet (Fellbaum, 1998; Ciaramita and Altun, 2006) . There are 41 supersense types: 26 for nouns and 15 for verbs, e.g., NOUN.ANIMAL and VERB.MOTION. For example, in NOUN.ANIMAL the relative frequency of human is 0.06, of dog is 0.01, of bird is 0.01, of cattle is 0.009, and of bumblebee is 0.0002.",
"cite_spans": [
{
"start": 166,
"end": 182,
"text": "(Fellbaum, 1998;",
"ref_id": null
},
{
"start": 183,
"end": 209,
"text": "Ciaramita and Altun, 2006)",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "\u2022 Relative frequency in a synset was calculated similarly to the previous feature category, but word frequencies were marginalized over Word-Net synsets (more fine-grained synonym sets). For example, in the synset {vet, warhorse, veteran, oldtimer, seasoned stager}, veteran is the most prototypical word, scoring 0.87.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Distributional and Linguistic Features",
"sec_num": "2.2"
},
{
"text": "We evaluate the utility of the pretrained word embeddings as features in downstream NLP tasks. We choose the following off-the-shelf models that utilize pretrained word embeddings as features:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation Benchmarks",
"sec_num": "3"
},
{
"text": "Sentiment Analysis (Senti). Socher et al. (2013) created a treebank of sentences annotated with fine-grained sentiment labels on phrases and sentences from movie review excerpts. The coarse-grained treebank of positive and negative classes has been split into training, development, and test datasets containing 6,920, 872, and 1,821 sentences, respectively. We use the average of the word vectors of a given sentence as a feature vector for classification (Faruqui et al., 2015; Sedoc et al., 2016) . The 2 -regularized logistic regression classifier is tuned on the development set and accuracy is reported on the test set.",
"cite_spans": [
{
"start": 457,
"end": 479,
"text": "(Faruqui et al., 2015;",
"ref_id": "BIBREF9"
},
{
"start": 480,
"end": 499,
"text": "Sedoc et al., 2016)",
"ref_id": "BIBREF36"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation Benchmarks",
"sec_num": "3"
},
{
"text": "Named Entity Recognition (NER). Named entity recognition is the task of identifying proper names in a sentence, such as names of persons, locations etc. We use the recently proposed LSTM-CRF NER model (Lample et al., 2016) which trains a forward-backward LSTM on a given sequence of words (represented as word vectors), the hidden units of which are then used as (the only) features in a CRF model (Lafferty et al., 2001) to predict the output label sequence. We use the CoNLL 2003 English NER dataset (Tjong Kim Sang and De Meulder, 2003) to train our models and present results on the test set.",
"cite_spans": [
{
"start": 201,
"end": 222,
"text": "(Lample et al., 2016)",
"ref_id": "BIBREF23"
},
{
"start": 398,
"end": 421,
"text": "(Lafferty et al., 2001)",
"ref_id": "BIBREF22"
},
{
"start": 513,
"end": 539,
"text": "Sang and De Meulder, 2003)",
"ref_id": "BIBREF47"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation Benchmarks",
"sec_num": "3"
},
{
"text": "Part of Speech Tagging (POS). For POS tagging, we again use the LSTM-CRF model (Lample et al., 2016), but instead of predicting the named entity tag for every word in a sentence, we train the tagger to predict the POS tag of the word. The tagger is trained and evaluated with the standard Penn TreeBank (PTB) (Marcus et al., 1993) training, development and test set splits as described in Collins (2002) .",
"cite_spans": [
{
"start": 309,
"end": 330,
"text": "(Marcus et al., 1993)",
"ref_id": "BIBREF26"
},
{
"start": 389,
"end": 403,
"text": "Collins (2002)",
"ref_id": "BIBREF6"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation Benchmarks",
"sec_num": "3"
},
{
"text": "Dependency Parsing (Parse). Dependency parsing is the task of identifying syntactic relations between the words of a sentence. For dependency parsing, we train the stack-LSTM parser of for English on the universal dependencies v1.1 treebank (Agi\u0107 et al., 2015) with the standard development and test splits, reporting unlabeled attachment scores (UAS) on the test data. We remove all part-of-speech and morphology features from the data, and prevent the model from optimizing the word embeddings used to represent each word in the corpus, thereby forcing the parser to rely completely on the pretrained embeddings.",
"cite_spans": [
{
"start": 241,
"end": 260,
"text": "(Agi\u0107 et al., 2015)",
"ref_id": "BIBREF0"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation Benchmarks",
"sec_num": "3"
},
{
"text": "Data. All models were trained on Wikipedia articles, split to paragraph-per-line. Texts were cleaned, tokenized, numbers were normalized by replacing each digit with \"DG\", all types that occur less than 10 times were replaces by the \"UNK\" token, the data was not lowercased. We list data sizes in table 1.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "4"
},
{
"text": "# paragraphs # tokens # types 2,532,361 100,872,713 156,663 Setup. 100-dimensional word embeddings were trained using the cbow model implemented in the word2vec toolkit (Mikolov et al., 2013) . 3 All training data was used, either shuffled or ordered by a curriculum. As described in \u00a73, we modified the extrinsic tasks to learn solely from word embeddings, without additional features. All models were learned under same conditions, across curricula: in Parse, NER, and POS we limited the number of training iterations to 3, 3, and 1, respectively. This setup allowed us to evaluate the effect of curriculum without additional interacting factors.",
"cite_spans": [
{
"start": 169,
"end": 191,
"text": "(Mikolov et al., 2013)",
"ref_id": "BIBREF27"
},
{
"start": 194,
"end": 195,
"text": "3",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "4"
},
{
"text": "Experiments. In all the experiments we first train word embedding models, then the word embeddings are used as features in four extrinsic tasks ( \u00a73). We tune the tasks on development data, and report results on the test data. The only component that varies across the experiments is order of paragraphs in the training corpus-the curriculum. We compare the following experimental setups:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "4"
},
{
"text": "\u2022 Shuffled baselines: the curriculum is defined by random shuffling the training data. We shuffled the data 10 times, and trained 10 word embeddings models, each model was then evaluated on downstream tasks. Following Bengio et al. (2009) , we report test results for the system that is closest to the median in dev scores. To evaluate variability and a range of scores that can be obtained from shuffling the data, we also report test results for systems that obtained the highest dev scores. \u2022 Sorted baselines: the curriculum is defined by sorting the training data by sentence length in increasing/decreasing order, similarly to (Spitkovsky et al., 2010). \u2022 Coherent baselines: the curriculum is defined by just concatenating Wikipedia articles. The goal of this experiment is to evaluate the importance of semantic coherence in training data.",
"cite_spans": [
{
"start": 218,
"end": 238,
"text": "Bengio et al. (2009)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "4"
},
{
"text": "Our intuition is that a coherent curriculum can improve models, since words with similar meanings and similar contexts are grouped when presented to the learner. \u2022 Optimized curriculum models: the curriculum is optimized using the BayesOpt. We evaluate and compare models optimized using features from one of the three feature groups ( \u00a72.2). As in the shuffled baselines, we fix the number of trials (here, BayesOpt iterations) to 10, and we report test results of systems that obtained best dev scores.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "4"
},
{
"text": "Results. Experimental results are listed in table 2. Most systems trained with curriculum substantially outperform the strongest of all baselines. These results are encouraging, given that all word embedding models were trained on the same set of examples, only in different order, and display the indirect influence of the data curriculum on downstream tasks. These results support our assumption that curriculum matters. Albeit not as pronounced as with optimized curriculum, sorting paragraphs by length can also lead to substantial improvements over random baselines, but there is no clear recipe on whether the models prefer curricula sorted in an increasing or decreasing order. These results also support the advantage of a taskspecific optimization framework over a general, intuition-guided recipe. An interesting result, also, that shuffling is not essential: systems trained on coherent data are on par (or better) than the shuffled systems. 4 In the next section, we analyze these results qualitatively.",
"cite_spans": [
{
"start": 953,
"end": 954,
"text": "4",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "4"
},
{
"text": "What are task-specific curriculum preferences? We manually inspect learned features and curriculum-sorted corpora, and find that best systems are obtained when their embeddings are learned from curricula appropriate to the downstream tasks. We discuss below several examples. POS and Parse systems converge to the same set of weights, when trained on features that provide various measures of syntactic simplicity. The features with highest coefficients (and thus the most important features in sorting) are #N P s, Parse tree depth, #V P s, and #P P s (in this order). The sign in the #N P s feature weight, however, is the opposite from the other three feature weights (i.e., sorted in different order). #N P s is sorted in the increasing order of the number of noun phrases in a paragraph, and the other features are sorted in the decreasing order. Since Wikipedia corpus contains a lot of partial phrases (titles and headings), such curriculum promotes more complex, full sentences, and demotes partial sentences.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "Best Senti system is sorted by prototypicality features. Most important features (with the highest coefficients) are Concreteness, Relative frequency in a supersense, and the Number of syllables. First two are sorted in decreasing order (i.e. paragraphs are sorted from more to less concrete, and from more to less prototypical words), and the Number of syllables is sorted in increasing order (this also promotes simpler, shorter words which are more prototypical). We hypothesize that this soring reflects the type of data that Sentiment analysis task is trained on: it is trained on movie reviews, that are usually written in a simple, colloquial language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "Unlike POS, Parse, and Senti systems, all NER systems prefer curricula in which texts are sorted from short to long paragraphs. The most important features in the best (simplicity-sorted) system are #P P s and Verb-token ratio, both sorted from less to more occurrences of prepositional and verb phrases. Interestingly, most of the top lines in the NER system curricula contain named entities, although none of our features mark named entities explicitly. We show top lines in the simplicityoptimized system in figure 2.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "Finally, in all systems sorted by prototypicality, the last line is indeed not a prototypical word Donaudampfschiffahrtselektrizit\u00e4tenhauptbetriebswerkbauunterbeamtengesellschaft, which is an actual word in German, frequently used as an example of compounding in synthetic languages, but rarely (or never?) used by German speakers.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "Weighting examples according to curriculum. Another way to integrate curriculum in word em- bedding training is to weight training examples according to curriculum during word representation training. We modify the cbow objective ( 1 1 + e \u2212weight (wt) ",
"cite_spans": [
{
"start": 248,
"end": 252,
"text": "(wt)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "+ \u03bb) log p(w t |w t\u2212c ..w t+c )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "Here, weight(w t ) denotes the score attributed to the token w t , which is the z-normalized score of the paragraph; \u03bb=0.5 is determined empirically. log p(w t )|w t\u2212c ..w t+c ) computes the probability of predicting word w t , using the context of c words to the left and right of w t . Notice that this quantity is no longer a proper probability, as we are not normalizing over the weights weight(w t ) over all tokens. However, the optimization in word2vec is performed using stochastic gradient descent, optimizing for a single token at each iteration. This yields a normalizer of 1 for each iteration, yielding the same gradient as the original cbow model.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "We retrain our best curriculum-sorted systems with the modified objective, also controlling for curriculum. The results are shown in table 3. We find that the benefit of integrating curriculum in training objective of word representations is not evident across tasks: Senti and NER systems trained on vectors with the modified objective substantially outperform best results in table 2; POS and Parse perform better than the baselines but worse than the systems with the original objective. Table 3 : Evaluation of the impact of curriculum integrated in the cbow objective.",
"cite_spans": [],
"ref_spans": [
{
"start": 491,
"end": 498,
"text": "Table 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Analysis",
"sec_num": "5"
},
{
"text": "Are we learning task-specific curricula? One way to assess whether we learn meaningful taskspecific curriculum preferences is to compare curricula learned by one downstream task across different feature groups. If learned curricula are similar in, say, NER system, despite being optimized once using diversity features and once using prototypicality features-two disjoint feature sets-we can infer that the NER task prefers word embeddings learned from examples presented in a certain order, regardless of specific optimization features. For each downstream task, we thus measure Spearman's rank correlation between the curricula optimized using diversity (D), or prototypicality (P), or simplicity (S) feature sets. Prior to measuring correlations, we remove duplicate lines from the training corpora. Correlation results across tasks and across feature sets are shown in table 4.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Senti NER POS",
"sec_num": null
},
{
"text": "The general pattern of results is that if two systems score higher than baselines, training sentences of their feature embeddings have similar curricula (i.e., the Spearman's \u03c1 is positive), and if two systems disagree (one is above and one is below the baseline), then their curricula also disagree (i.e., the Spearman's \u03c1 is negative or close to zero). NER systems all outperform the baselines and their curricula have high correlations. Moreover, NER sorted by diversity and simplicity have better scores than NER sorted by prototypicality, and in line with these results \u03c1(S,D) N ER > \u03c1(P,S) N ER and \u03c1(S,D) N ER > \u03c1(D,P) N ER . Similar pattern of results is in POS correlations. In Parse systems, also, diversity and simplicity features yielded best parsing results, and \u03c1(S,D) P arse has high positive correlation. The prototypicality-optimized parser performed poorly, and its correlations with better systems are negative. The best parser was trained using the diversity-optimized curriculum, and thus \u03c1(D,P) P arse is the lowest. Senti results follow similar pattern of curricula correlations. Table 4 : Curricula correlations across feature groups.",
"cite_spans": [],
"ref_spans": [
{
"start": 1103,
"end": 1110,
"text": "Table 4",
"ref_id": null
}
],
"eq_spans": [],
"section": "Senti NER POS",
"sec_num": null
},
{
"text": "Curriculum learning vs. data selection. We compare the task of curriculum learning to the task of data selection (reducing the set of training instances to more important or cleaner examples). We reduce the training data to the subset of 10% of tokens, and train downstream tasks on the reduced training sets. We compare system performance trained using the top 10% of tokens in the best curriculum-sorted systems (Senti-prototypicality, NER-implicity, POS-simplicity, Parse-diversity) to the systems trained using the top 10% of tokens in a corpus with randomly shuffled paragraphs. 6 The results are listed in table 5. The curriculum-based systems are better in POS 6 Top n% tokens are used rather than top n% paragraphs because in all tasks except NER curriculum-sorted corpora begin with longer paragraphs. Thus, with top n% paragraphs our systems would have an advantage over random systems due to larger vocabulary sizes and not necessarily due to a better subset of data. Table 5 : Data selection results. and in Parse systems, mainly because these tasks prefer vectors trained on curricula that promote well-formed sentences (as discussed above). Conversely, NER prefers vectors trained on corpora that begin with named entities, so most of the tokens in the reduced training data are constituents in short noun phrases. These results suggest that the tasks of data selection and curriculum learning are different. Curriculum is about strong initialization of the models and time-course learning, which is not necessarily sufficient for data reduction.",
"cite_spans": [
{
"start": 584,
"end": 585,
"text": "6",
"ref_id": null
},
{
"start": 668,
"end": 669,
"text": "6",
"ref_id": null
}
],
"ref_spans": [
{
"start": 979,
"end": 986,
"text": "Table 5",
"ref_id": null
}
],
"eq_spans": [],
"section": "Senti NER POS",
"sec_num": null
},
{
"text": "Two prior studies on curriculum learning in NLP are discussed in the paper (Bengio et al., 2009; Spitkovsky et al., 2010) . Curriculum learning and related research on self-paced learning has been explored more deeply in computer vision (Bengio et al., 2009; Kumar et al., 2010; Lee and Grauman, 2011) and in multimedia analysis (Jiang et al., 2015) . Bayesian optimization has also received little attention in NLP. GPs were used in the task of machine translation quality estimation (Cohn and Specia, 2013) and in temporal analysis of social media texts (Preotiuc-Pietro and Cohn, 2013) ; TPEs were used by Yogatama et al. (2015) for optimizing choices of feature representations-ngram size, regularization choice, etc.-in supervised classifiers.",
"cite_spans": [
{
"start": 75,
"end": 96,
"text": "(Bengio et al., 2009;",
"ref_id": "BIBREF1"
},
{
"start": 97,
"end": 121,
"text": "Spitkovsky et al., 2010)",
"ref_id": "BIBREF42"
},
{
"start": 237,
"end": 258,
"text": "(Bengio et al., 2009;",
"ref_id": "BIBREF1"
},
{
"start": 259,
"end": 278,
"text": "Kumar et al., 2010;",
"ref_id": "BIBREF19"
},
{
"start": 279,
"end": 301,
"text": "Lee and Grauman, 2011)",
"ref_id": "BIBREF24"
},
{
"start": 329,
"end": 349,
"text": "(Jiang et al., 2015)",
"ref_id": "BIBREF16"
},
{
"start": 485,
"end": 508,
"text": "(Cohn and Specia, 2013)",
"ref_id": "BIBREF5"
},
{
"start": 556,
"end": 588,
"text": "(Preotiuc-Pietro and Cohn, 2013)",
"ref_id": "BIBREF30"
},
{
"start": 609,
"end": 631,
"text": "Yogatama et al. (2015)",
"ref_id": "BIBREF51"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "6"
},
{
"text": "We used Bayesian optimization to optimize curricula for training dense distributed word representations, which, in turn, were used as the sole features in NLP tasks. Our experiments confirmed that better curricula yield stronger models. We also conducted an extensive analysis, which sheds better light on understanding of text properties that are beneficial for model initialization. The proposed novel technique for finding an optimal curriculum is general, and can be used with other datasets and models.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "7"
},
{
"text": "Intuitively, this feature promotes paragraphs that contain semantically similar high-probability words.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "http://http://people.sutd.edu.sg/ yue_zhang/doc",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "To evaluate the impact of curriculum learning, we enforced sequential processing of data organized in a predefined order of training examples. To control for sequential processing, word embedding were learned by running the cbow using a single thread for one iteration.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "Note that in the shuffled NER baselines, best dev results yield lower performance on the test data. This implies that in the standard development/test splits the development and test sets are not fully compatible or not large enough. We also observe this problem in the curriculum-optimized Parse-prototypicality and Senti-diversity systems. The dev scores for the Parse systems are 76.99, 76.47, 76.47 for diversity, prototypicality, and simplicity, respectively, but the prototypicality-sorted parser performs poorly on test data. Similarly in the sentiment analysis task, the dev scores are 69.15, 69.04, 69.49 for diversity, prototypicality, and simplicity feature groups. Senti-diversity scores, however, are lower on the test data, although the dev results are better than in Senti-simplicity. This limitation of the standard dev/test splits is beyond the scope of this paper.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "The modified word2vec tool is located at https:// github.com/wlin12/wang2vec .",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "This work was supported by the National Science Foundation through award IIS-1526745. We are grateful to Nathan Schneider, Guillaume Lample, Waleed Ammar, Austin Matthews, and the anonymous reviewers for their insightful comments.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgments",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Prokopis Prokopidis, Sampo Pyysalo, Wolfgang Seeker",
"authors": [
{
"first": "\u017deljko",
"middle": [],
"last": "Agi\u0107",
"suffix": ""
},
{
"first": "Maria",
"middle": [
"Jesus"
],
"last": "Aranzabe",
"suffix": ""
},
{
"first": "Aitziber",
"middle": [],
"last": "Atutxa",
"suffix": ""
},
{
"first": "Cristina",
"middle": [],
"last": "Bosco",
"suffix": ""
},
{
"first": "Jinho",
"middle": [],
"last": "Choi",
"suffix": ""
},
{
"first": "Marie-Catherine",
"middle": [],
"last": "De Marneffe",
"suffix": ""
},
{
"first": "Timothy",
"middle": [],
"last": "Dozat",
"suffix": ""
},
{
"first": "Rich\u00e1rd",
"middle": [],
"last": "Farkas",
"suffix": ""
},
{
"first": "Jennifer",
"middle": [],
"last": "Foster",
"suffix": ""
},
{
"first": "Filip",
"middle": [],
"last": "Ginter",
"suffix": ""
},
{
"first": "Iakes",
"middle": [],
"last": "Goenaga",
"suffix": ""
},
{
"first": "Koldo",
"middle": [],
"last": "Gojenola",
"suffix": ""
},
{
"first": "Yoav",
"middle": [],
"last": "Goldberg",
"suffix": ""
},
{
"first": "Jan",
"middle": [],
"last": "Haji\u010d",
"suffix": ""
},
{
"first": "Anders",
"middle": [
"Traerup"
],
"last": "Johannsen",
"suffix": ""
},
{
"first": "Jenna",
"middle": [],
"last": "Kanerva",
"suffix": ""
},
{
"first": "Juha",
"middle": [],
"last": "Kuokkala",
"suffix": ""
},
{
"first": "Veronika",
"middle": [],
"last": "Laippala",
"suffix": ""
},
{
"first": "Alessandro",
"middle": [],
"last": "Lenci",
"suffix": ""
},
{
"first": "Krister",
"middle": [],
"last": "Lind\u00e9n",
"suffix": ""
},
{
"first": "Nikola",
"middle": [],
"last": "Ljube\u0161i\u0107",
"suffix": ""
}
],
"year": 2015,
"venue": "Mojgan Seraji",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "\u017deljko Agi\u0107, Maria Jesus Aranzabe, Aitziber Atutxa, Cristina Bosco, Jinho Choi, Marie-Catherine de Marneffe, Timothy Dozat, Rich\u00e1rd Farkas, Jennifer Foster, Filip Ginter, Iakes Goenaga, Koldo Gojenola, Yoav Goldberg, Jan Haji\u010d, An- ders Traerup Johannsen, Jenna Kanerva, Juha Kuokkala, Veronika Laippala, Alessandro Lenci, Krister Lind\u00e9n, Nikola Ljube\u0161i\u0107, Teresa Lynn, Christopher Manning, H\u00e9ctor Alonso Mart\u00ednez, Ryan McDonald, Anna Missil\u00e4, Simonetta Monte- magni, Joakim Nivre, Hanna Nurmi, Petya Osen- ova, Slav Petrov, Jussi Piitulainen, Barbara Plank, Prokopis Prokopidis, Sampo Pyysalo, Wolfgang Seeker, Mojgan Seraji, Natalia Silveira, Maria Simi, Kiril Simov, Aaron Smith, Reut Tsarfaty, Veronika Vincze, and Daniel Zeman. 2015. Universal de- pendencies 1.1. LINDAT/CLARIN digital library at Institute of Formal and Applied Linguistics, Charles University in Prague.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Curriculum learning",
"authors": [
{
"first": "Yoshua",
"middle": [],
"last": "Bengio",
"suffix": ""
},
{
"first": "J\u00e9r\u00f4me",
"middle": [],
"last": "Louradour",
"suffix": ""
},
{
"first": "Ronan",
"middle": [],
"last": "Collobert",
"suffix": ""
},
{
"first": "Jason",
"middle": [],
"last": "Weston",
"suffix": ""
}
],
"year": 2009,
"venue": "Proc. ICML",
"volume": "",
"issue": "",
"pages": "41--48",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yoshua Bengio, J\u00e9r\u00f4me Louradour, Ronan Collobert, and Jason Weston. 2009. Curriculum learning. In Proc. ICML, pages 41-48.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Algorithms for hyper-parameter optimization",
"authors": [
{
"first": "S",
"middle": [],
"last": "James",
"suffix": ""
},
{
"first": "R\u00e9mi",
"middle": [],
"last": "Bergstra",
"suffix": ""
},
{
"first": "Yoshua",
"middle": [],
"last": "Bardenet",
"suffix": ""
},
{
"first": "Bal\u00e1zs",
"middle": [],
"last": "Bengio",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "K\u00e9gl",
"suffix": ""
}
],
"year": 2011,
"venue": "Proc. NIPS",
"volume": "",
"issue": "",
"pages": "2546--2554",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "James S Bergstra, R\u00e9mi Bardenet, Yoshua Bengio, and Bal\u00e1zs K\u00e9gl. 2011. Algorithms for hyper-parameter optimization. In Proc. NIPS, pages 2546-2554.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Concreteness ratings for 40 thousand generally known english word lemmas. Behavior research methods",
"authors": [
{
"first": "Marc",
"middle": [],
"last": "Brysbaert",
"suffix": ""
},
{
"first": "Amy",
"middle": [
"Beth"
],
"last": "Warriner",
"suffix": ""
},
{
"first": "Victor",
"middle": [],
"last": "Kuperman",
"suffix": ""
}
],
"year": 2014,
"venue": "",
"volume": "46",
"issue": "",
"pages": "904--911",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Marc Brysbaert, Amy Beth Warriner, and Victor Ku- perman. 2014. Concreteness ratings for 40 thou- sand generally known english word lemmas. Behav- ior research methods, 46(3):904-911.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Broad-coverage sense disambiguation and information extraction with a supersense sequence tagger",
"authors": [
{
"first": "Massimiliano",
"middle": [],
"last": "Ciaramita",
"suffix": ""
},
{
"first": "Yasemin",
"middle": [],
"last": "Altun",
"suffix": ""
}
],
"year": 2006,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "594--602",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Massimiliano Ciaramita and Yasemin Altun. 2006. Broad-coverage sense disambiguation and informa- tion extraction with a supersense sequence tagger. In Proc. EMNLP, pages 594-602.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Modelling annotator bias with multi-task Gaussian processes: An application to machine translation quality estimation",
"authors": [
{
"first": "Trevor",
"middle": [],
"last": "Cohn",
"suffix": ""
},
{
"first": "Lucia",
"middle": [],
"last": "Specia",
"suffix": ""
}
],
"year": 2013,
"venue": "Proc. ACL",
"volume": "",
"issue": "",
"pages": "32--42",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Trevor Cohn and Lucia Specia. 2013. Modelling an- notator bias with multi-task Gaussian processes: An application to machine translation quality estima- tion. In Proc. ACL, pages 32-42.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "Discriminative training methods for hidden markov models: Theory and experiments with perceptron algorithms",
"authors": [
{
"first": "Michael",
"middle": [],
"last": "Collins",
"suffix": ""
}
],
"year": 2002,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "1--8",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Michael Collins. 2002. Discriminative training meth- ods for hidden markov models: Theory and experi- ments with perceptron algorithms. In Proc. EMNLP, pages 1-8.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Transitionbased dependency parsing with stack long shortterm memory",
"authors": [
{
"first": "Chris",
"middle": [],
"last": "Dyer",
"suffix": ""
},
{
"first": "Miguel",
"middle": [],
"last": "Ballesteros",
"suffix": ""
},
{
"first": "Wang",
"middle": [],
"last": "Ling",
"suffix": ""
},
{
"first": "Austin",
"middle": [],
"last": "Matthews",
"suffix": ""
},
{
"first": "Noah",
"middle": [
"A"
],
"last": "Smith",
"suffix": ""
}
],
"year": 2015,
"venue": "Proc. ACL",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Chris Dyer, Miguel Ballesteros, Wang Ling, Austin Matthews, and Noah A. Smith. 2015. Transition- based dependency parsing with stack long short- term memory. In Proc. ACL.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Learning and development in neural networks: The importance of starting small",
"authors": [
{
"first": "",
"middle": [],
"last": "Jeffrey L Elman",
"suffix": ""
}
],
"year": 1993,
"venue": "Cognition",
"volume": "48",
"issue": "1",
"pages": "71--99",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Jeffrey L Elman. 1993. Learning and development in neural networks: The importance of starting small. Cognition, 48(1):71-99.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Retrofitting word vectors to semantic lexicons",
"authors": [
{
"first": "Manaal",
"middle": [],
"last": "Faruqui",
"suffix": ""
},
{
"first": "Jesse",
"middle": [],
"last": "Dodge",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Sujay",
"suffix": ""
},
{
"first": "Chris",
"middle": [],
"last": "Jauhar",
"suffix": ""
},
{
"first": "Eduard",
"middle": [],
"last": "Dyer",
"suffix": ""
},
{
"first": "Noah",
"middle": [
"A"
],
"last": "Hovy",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Smith",
"suffix": ""
}
],
"year": 2015,
"venue": "Proc. NAACL",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Manaal Faruqui, Jesse Dodge, Sujay K. Jauhar, Chris Dyer, Eduard Hovy, and Noah A. Smith. 2015. Retrofitting word vectors to semantic lexicons. In Proc. NAACL.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "WordNet: an electronic lexical database",
"authors": [],
"year": 1998,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Christiane Fellbaum, editor. 1998. WordNet: an elec- tronic lexical database. MIT Press.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "A systematic exploration of diversity in machine translation",
"authors": [
{
"first": "Kevin",
"middle": [],
"last": "Gimpel",
"suffix": ""
},
{
"first": "Dhruv",
"middle": [],
"last": "Batra",
"suffix": ""
},
{
"first": "Chris",
"middle": [],
"last": "Dyer",
"suffix": ""
},
{
"first": "Gregory",
"middle": [],
"last": "Shakhnarovich",
"suffix": ""
}
],
"year": 2013,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "1100--1111",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kevin Gimpel, Dhruv Batra, Chris Dyer, and Gregory Shakhnarovich. 2013. A systematic exploration of diversity in machine translation. In Proc. EMNLP, pages 1100-1111.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Combining lexical and grammatical features to improve readability measures for first and second language texts",
"authors": [
{
"first": "J",
"middle": [],
"last": "Michael",
"suffix": ""
},
{
"first": "Kevyn",
"middle": [],
"last": "Heilman",
"suffix": ""
},
{
"first": "Jamie",
"middle": [],
"last": "Collins-Thompson",
"suffix": ""
},
{
"first": "Maxine",
"middle": [],
"last": "Callan",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Eskenazi",
"suffix": ""
}
],
"year": 2007,
"venue": "Proc. NAACL",
"volume": "",
"issue": "",
"pages": "460--467",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Michael J. Heilman, Kevyn Collins-Thompson, Jamie Callan, and Maxine Eskenazi. 2007. Combining lexical and grammatical features to improve read- ability measures for first and second language texts. In Proc. NAACL, pages 460-467.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Entropy search for information-efficient global optimization",
"authors": [
{
"first": "Philipp",
"middle": [],
"last": "Hennig",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Christian",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Schuler",
"suffix": ""
}
],
"year": 2012,
"venue": "The Journal of Machine Learning Research",
"volume": "13",
"issue": "1",
"pages": "1809--1837",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Philipp Hennig and Christian J Schuler. 2012. En- tropy search for information-efficient global opti- mization. The Journal of Machine Learning Re- search, 13(1):1809-1837.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Portfolio allocation for Bayesian optimization",
"authors": [
{
"first": "Eric",
"middle": [],
"last": "Matthew D Hoffman",
"suffix": ""
},
{
"first": "Nando",
"middle": [],
"last": "Brochu",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "De Freitas",
"suffix": ""
}
],
"year": 2011,
"venue": "Proc. UAI",
"volume": "",
"issue": "",
"pages": "327--336",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Matthew D Hoffman, Eric Brochu, and Nando de Fre- itas. 2011. Portfolio allocation for Bayesian opti- mization. In Proc. UAI, pages 327-336.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Self-paced learning with diversity",
"authors": [
{
"first": "Lu",
"middle": [],
"last": "Jiang",
"suffix": ""
},
{
"first": "Deyu",
"middle": [],
"last": "Meng",
"suffix": ""
},
{
"first": "-I",
"middle": [],
"last": "Shoou",
"suffix": ""
},
{
"first": "Zhenzhong",
"middle": [],
"last": "Yu",
"suffix": ""
},
{
"first": "Shiguang",
"middle": [],
"last": "Lan",
"suffix": ""
},
{
"first": "Alexander",
"middle": [],
"last": "Shan",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Hauptmann",
"suffix": ""
}
],
"year": 2014,
"venue": "Proc. NIPS",
"volume": "",
"issue": "",
"pages": "2078--2086",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lu Jiang, Deyu Meng, Shoou-I Yu, Zhenzhong Lan, Shiguang Shan, and Alexander Hauptmann. 2014. Self-paced learning with diversity. In Proc. NIPS, pages 2078-2086.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Self-paced curriculum learning",
"authors": [
{
"first": "Lu",
"middle": [],
"last": "Jiang",
"suffix": ""
},
{
"first": "Deyu",
"middle": [],
"last": "Meng",
"suffix": ""
},
{
"first": "Qian",
"middle": [],
"last": "Zhao",
"suffix": ""
},
{
"first": "Shiguang",
"middle": [],
"last": "Shan",
"suffix": ""
},
{
"first": "Alexander",
"middle": [
"G"
],
"last": "Hauptmann",
"suffix": ""
}
],
"year": 2015,
"venue": "Proc. AAAI",
"volume": "2",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lu Jiang, Deyu Meng, Qian Zhao, Shiguang Shan, and Alexander G Hauptmann. 2015. Self-paced curricu- lum learning. In Proc. AAAI, volume 2, page 6.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "A taxonomy of global optimization methods based on response surfaces",
"authors": [
{
"first": "R",
"middle": [],
"last": "Donald",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Jones",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "21",
"issue": "",
"pages": "345--383",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Donald R Jones. 2001. A taxonomy of global opti- mization methods based on response surfaces. Jour- nal of global optimization, 21(4):345-383.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "The development of memory in children",
"authors": [
{
"first": "Robert",
"middle": [],
"last": "Kail",
"suffix": ""
}
],
"year": 1990,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Robert Kail. 1990. The development of memory in children. W. H. Freeman and Company, 3rd edition.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "Self-paced learning for latent variable models",
"authors": [
{
"first": "M",
"middle": [
"P"
],
"last": "Kumar",
"suffix": ""
},
{
"first": "Benjamin",
"middle": [],
"last": "Packer",
"suffix": ""
},
{
"first": "Daphne",
"middle": [],
"last": "Koller",
"suffix": ""
}
],
"year": 2010,
"venue": "Proc. NIPS",
"volume": "",
"issue": "",
"pages": "1189--1197",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. P. Kumar, Benjamin Packer, and Daphne Koller. 2010. Self-paced learning for latent variable mod- els. In Proc. NIPS, pages 1189-1197.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Age-of-acquisition ratings for 30,000 english words",
"authors": [
{
"first": "Victor",
"middle": [],
"last": "Kuperman",
"suffix": ""
},
{
"first": "Hans",
"middle": [],
"last": "Stadthagen-Gonzalez",
"suffix": ""
},
{
"first": "Marc",
"middle": [],
"last": "Brysbaert",
"suffix": ""
}
],
"year": 2012,
"venue": "",
"volume": "44",
"issue": "",
"pages": "978--990",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Victor Kuperman, Hans Stadthagen-Gonzalez, and Marc Brysbaert. 2012. Age-of-acquisition ratings for 30,000 english words. Behavior Research Meth- ods, 44(4):978-990.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise",
"authors": [
{
"first": "J",
"middle": [],
"last": "Harold",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Kushner",
"suffix": ""
}
],
"year": 1964,
"venue": "Journal of Basic Engineering",
"volume": "86",
"issue": "1",
"pages": "97--106",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Harold J Kushner. 1964. A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise. Journal of Basic Engineering, 86(1):97-106.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Conditional random fields: Probabilistic models for segmenting and labeling sequence data",
"authors": [
{
"first": "John",
"middle": [],
"last": "Lafferty",
"suffix": ""
},
{
"first": "Andrew",
"middle": [],
"last": "Mccallum",
"suffix": ""
},
{
"first": "Fernando",
"middle": [],
"last": "Pereira",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "282--289",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "John Lafferty, Andrew McCallum, and Fernando Pereira. 2001. Conditional random fields: Prob- abilistic models for segmenting and labeling se- quence data. pages 282-289.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "Neural architectures for named entity recognition",
"authors": [
{
"first": "Guillaume",
"middle": [],
"last": "Lample",
"suffix": ""
},
{
"first": "Miguel",
"middle": [],
"last": "Ballesteros",
"suffix": ""
},
{
"first": "Sandeep",
"middle": [],
"last": "Subramanian",
"suffix": ""
},
{
"first": "Kazuya",
"middle": [],
"last": "Kawakami",
"suffix": ""
},
{
"first": "Chris",
"middle": [],
"last": "Dyer",
"suffix": ""
}
],
"year": 2016,
"venue": "Proc. NAACL",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Guillaume Lample, Miguel Ballesteros, Sandeep Sub- ramanian, Kazuya Kawakami, and Chris Dyer. 2016. Neural architectures for named entity recog- nition. In Proc. NAACL.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "Learning the easy things first: Self-paced visual category discovery",
"authors": [
{
"first": "Jae",
"middle": [],
"last": "Yong",
"suffix": ""
},
{
"first": "Kristen",
"middle": [],
"last": "Lee",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Grauman",
"suffix": ""
}
],
"year": 2011,
"venue": "Proc. CVPR",
"volume": "",
"issue": "",
"pages": "1721--1728",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yong Jae Lee and Kristen Grauman. 2011. Learning the easy things first: Self-paced visual category dis- covery. In Proc. CVPR, pages 1721-1728.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Measuring biological diversity",
"authors": [
{
"first": "Anne",
"middle": [
"E"
],
"last": "Magurran",
"suffix": ""
}
],
"year": 2013,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Anne E Magurran. 2013. Measuring biological diver- sity. John Wiley & Sons.",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Building a large annotated corpus of english: The penn treebank",
"authors": [
{
"first": "P",
"middle": [],
"last": "Mitchell",
"suffix": ""
},
{
"first": "Mary",
"middle": [
"Ann"
],
"last": "Marcus",
"suffix": ""
},
{
"first": "Beatrice",
"middle": [],
"last": "Marcinkiewicz",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Santorini",
"suffix": ""
}
],
"year": 1993,
"venue": "Computational linguistics",
"volume": "19",
"issue": "2",
"pages": "313--330",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Mitchell P Marcus, Mary Ann Marcinkiewicz, and Beatrice Santorini. 1993. Building a large anno- tated corpus of english: The penn treebank. Compu- tational linguistics, 19(2):313-330.",
"links": null
},
"BIBREF27": {
"ref_id": "b27",
"title": "Efficient estimation of word representations in vector space",
"authors": [
{
"first": "Tomas",
"middle": [],
"last": "Mikolov",
"suffix": ""
},
{
"first": "Kai",
"middle": [],
"last": "Chen",
"suffix": ""
},
{
"first": "Greg",
"middle": [],
"last": "Corrado",
"suffix": ""
},
{
"first": "Jeffrey",
"middle": [],
"last": "Dean",
"suffix": ""
}
],
"year": 2013,
"venue": "Proc. ICLR",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. Efficient estimation of word represen- tations in vector space. In Proc. ICLR.",
"links": null
},
"BIBREF28": {
"ref_id": "b28",
"title": "On Bayesian methods for seeking the extremum",
"authors": [
{
"first": "Jonas",
"middle": [],
"last": "Mo\u010dkus",
"suffix": ""
}
],
"year": 1978,
"venue": "Vytautas Tiesis, and Antanas \u017dilinskas",
"volume": "2",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Jonas Mo\u010dkus, Vytautas Tiesis, and Antanas \u017dilin- skas. 1978. On Bayesian methods for seeking the extremum. Towards global optimization, 2(117- 129):2.",
"links": null
},
"BIBREF29": {
"ref_id": "b29",
"title": "Revisiting readability: A unified framework for predicting text quality",
"authors": [
{
"first": "Emily",
"middle": [],
"last": "Pitler",
"suffix": ""
},
{
"first": "Ani",
"middle": [],
"last": "Nenkova",
"suffix": ""
}
],
"year": 2008,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "186--195",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Emily Pitler and Ani Nenkova. 2008. Revisiting readability: A unified framework for predicting text quality. In Proc. EMNLP, pages 186-195.",
"links": null
},
"BIBREF30": {
"ref_id": "b30",
"title": "A temporal model of text periodicities using gaussian processes",
"authors": [
{
"first": "Daniel",
"middle": [],
"last": "Preotiuc",
"suffix": ""
},
{
"first": "-Pietro",
"middle": [],
"last": "",
"suffix": ""
},
{
"first": "Trevor",
"middle": [],
"last": "Cohn",
"suffix": ""
}
],
"year": 2013,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "977--988",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Daniel Preotiuc-Pietro and Trevor Cohn. 2013. A tem- poral model of text periodicities using gaussian pro- cesses. In Proc. EMNLP, pages 977-988.",
"links": null
},
"BIBREF31": {
"ref_id": "b31",
"title": "Diversity and dissimilarity coefficients: a unified approach",
"authors": [
{
"first": "",
"middle": [],
"last": "C Radhakrishna Rao",
"suffix": ""
}
],
"year": 1982,
"venue": "",
"volume": "21",
"issue": "",
"pages": "24--43",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "C Radhakrishna Rao. 1982. Diversity and dissimilarity coefficients: a unified approach. Theoretical popu- lation biology, 21(1):24-43.",
"links": null
},
"BIBREF32": {
"ref_id": "b32",
"title": "Gaussian Processes for machine learning",
"authors": [
{
"first": "Carl",
"middle": [
"Edward"
],
"last": "Rasmussen",
"suffix": ""
}
],
"year": 2006,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Carl Edward Rasmussen. 2006. Gaussian Processes for machine learning.",
"links": null
},
"BIBREF33": {
"ref_id": "b33",
"title": "Principles of categorization",
"authors": [
{
"first": "Eleanor",
"middle": [],
"last": "Rosch",
"suffix": ""
}
],
"year": 1978,
"venue": "Cognition and categorization",
"volume": "",
"issue": "",
"pages": "28--71",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Eleanor Rosch. 1978. Principles of categorization. In Eleanor Rosch and Barbara B. Lloyd, editors, Cog- nition and categorization, pages 28-71.",
"links": null
},
"BIBREF34": {
"ref_id": "b34",
"title": "Species diversity in space and time",
"authors": [
{
"first": "",
"middle": [],
"last": "Michael L Rosenzweig",
"suffix": ""
}
],
"year": 1995,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Michael L Rosenzweig. 1995. Species diversity in space and time. Cambridge University Press.",
"links": null
},
"BIBREF35": {
"ref_id": "b35",
"title": "Reading level assessment using support vector machines and statistical language models",
"authors": [
{
"first": "Sarah",
"middle": [
"E"
],
"last": "Schwarm",
"suffix": ""
},
{
"first": "Mari",
"middle": [],
"last": "Ostendorf",
"suffix": ""
}
],
"year": 2005,
"venue": "Proc. ACL",
"volume": "",
"issue": "",
"pages": "523--530",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sarah E. Schwarm and Mari Ostendorf. 2005. Read- ing level assessment using support vector machines and statistical language models. In Proc. ACL, pages 523-530.",
"links": null
},
"BIBREF36": {
"ref_id": "b36",
"title": "Semantic word clusters using signed normalized graph cuts",
"authors": [
{
"first": "Jo\u00e3o",
"middle": [],
"last": "Sedoc",
"suffix": ""
},
{
"first": "Jean",
"middle": [],
"last": "Gallier",
"suffix": ""
},
{
"first": "Lyle",
"middle": [],
"last": "Ungar",
"suffix": ""
},
{
"first": "Dean",
"middle": [],
"last": "Foster",
"suffix": ""
}
],
"year": 2016,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {
"arXiv": [
"arXiv:1601.05403"
]
},
"num": null,
"urls": [],
"raw_text": "Jo\u00e3o Sedoc, Jean Gallier, Lyle Ungar, and Dean Foster. 2016. Semantic word clusters using signed normal- ized graph cuts. arXiv preprint arXiv:1601.05403.",
"links": null
},
"BIBREF37": {
"ref_id": "b37",
"title": "Taking the human out of the loop: A review of Bayesian optimization",
"authors": [
{
"first": "Bobak",
"middle": [],
"last": "Shahriari",
"suffix": ""
},
{
"first": "Kevin",
"middle": [],
"last": "Swersky",
"suffix": ""
},
{
"first": "Ziyu",
"middle": [],
"last": "Wang",
"suffix": ""
},
{
"first": "P",
"middle": [],
"last": "Ryan",
"suffix": ""
},
{
"first": "Nando",
"middle": [],
"last": "Adams",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "De Freitas",
"suffix": ""
}
],
"year": 2016,
"venue": "Proc. IEEE",
"volume": "104",
"issue": "",
"pages": "148--175",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Bobak Shahriari, Kevin Swersky, Ziyu Wang, Ryan P Adams, and Nando de Freitas. 2016. Taking the human out of the loop: A review of Bayesian opti- mization. Proc. IEEE, 104(1):148-175.",
"links": null
},
"BIBREF38": {
"ref_id": "b38",
"title": "Measurement of diversity",
"authors": [
{
"first": "H",
"middle": [],
"last": "Edward",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Simpson",
"suffix": ""
}
],
"year": 1949,
"venue": "Nature",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Edward H Simpson. 1949. Measurement of diversity. Nature.",
"links": null
},
"BIBREF39": {
"ref_id": "b39",
"title": "The behavior of organisms: an experimental analysis. An Experimental Analysis",
"authors": [
{
"first": "Frederic",
"middle": [],
"last": "Burrhus",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Skinner",
"suffix": ""
}
],
"year": 1938,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Burrhus Frederic Skinner. 1938. The behavior of or- ganisms: an experimental analysis. An Experimen- tal Analysis.",
"links": null
},
"BIBREF40": {
"ref_id": "b40",
"title": "Practical Bayesian optimization of machine learning algorithms",
"authors": [
{
"first": "Jasper",
"middle": [],
"last": "Snoek",
"suffix": ""
},
{
"first": "Hugo",
"middle": [],
"last": "Larochelle",
"suffix": ""
},
{
"first": "Ryan P",
"middle": [],
"last": "Adams",
"suffix": ""
}
],
"year": 2012,
"venue": "Proc. NIPS",
"volume": "",
"issue": "",
"pages": "2951--2959",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Jasper Snoek, Hugo Larochelle, and Ryan P Adams. 2012. Practical Bayesian optimization of machine learning algorithms. In Proc. NIPS, pages 2951- 2959.",
"links": null
},
"BIBREF41": {
"ref_id": "b41",
"title": "Recursive deep models for semantic compositionality over a sentiment treebank",
"authors": [
{
"first": "Richard",
"middle": [],
"last": "Socher",
"suffix": ""
},
{
"first": "Alex",
"middle": [],
"last": "Perelygin",
"suffix": ""
},
{
"first": "Jean",
"middle": [],
"last": "Wu",
"suffix": ""
},
{
"first": "Jason",
"middle": [],
"last": "Chuang",
"suffix": ""
},
{
"first": "Christopher",
"middle": [
"D"
],
"last": "Manning",
"suffix": ""
},
{
"first": "Andrew",
"middle": [
"Y"
],
"last": "Ng",
"suffix": ""
},
{
"first": "Christopher",
"middle": [],
"last": "Potts",
"suffix": ""
}
],
"year": 2013,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D. Manning, Andrew Y. Ng, and Christopher Potts. 2013. Recursive deep mod- els for semantic compositionality over a sentiment treebank. In Proc. EMNLP.",
"links": null
},
"BIBREF42": {
"ref_id": "b42",
"title": "From baby steps to leapfrog: How less is more in unsupervised dependency parsing",
"authors": [
{
"first": "Hiyan",
"middle": [],
"last": "Valentin I Spitkovsky",
"suffix": ""
},
{
"first": "Dan",
"middle": [
"Jurafsky"
],
"last": "Alshawi",
"suffix": ""
}
],
"year": 2010,
"venue": "Proc. NAACL",
"volume": "",
"issue": "",
"pages": "751--759",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Valentin I Spitkovsky, Hiyan Alshawi, and Dan Juraf- sky. 2010. From baby steps to leapfrog: How less is more in unsupervised dependency parsing. In Proc. NAACL, pages 751-759.",
"links": null
},
"BIBREF43": {
"ref_id": "b43",
"title": "Gaussian process optimization in the bandit setting: No regret and experimental design",
"authors": [
{
"first": "Niranjan",
"middle": [],
"last": "Srinivas",
"suffix": ""
},
{
"first": "Andreas",
"middle": [],
"last": "Krause",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Sham",
"suffix": ""
},
{
"first": "Matthias",
"middle": [],
"last": "Kakade",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Seeger",
"suffix": ""
}
],
"year": 2010,
"venue": "Proc. ICML",
"volume": "",
"issue": "",
"pages": "1015--1022",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Niranjan Srinivas, Andreas Krause, Sham M Kakade, and Matthias Seeger. 2010. Gaussian process opti- mization in the bandit setting: No regret and experi- mental design. In Proc. ICML, pages 1015-1022.",
"links": null
},
"BIBREF44": {
"ref_id": "b44",
"title": "A general framework for analysing diversity in science, technology and society",
"authors": [
{
"first": "Andy",
"middle": [],
"last": "Stirling",
"suffix": ""
}
],
"year": 2007,
"venue": "Journal of the Royal Society Interface",
"volume": "4",
"issue": "15",
"pages": "707--719",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Andy Stirling. 2007. A general framework for analysing diversity in science, technology and so- ciety. Journal of the Royal Society Interface, 4(15):707-719.",
"links": null
},
"BIBREF45": {
"ref_id": "b45",
"title": "An analysis of diversity measures",
"authors": [
{
"first": "",
"middle": [],
"last": "E Ke Tang",
"suffix": ""
},
{
"first": "N",
"middle": [],
"last": "Ponnuthurai",
"suffix": ""
},
{
"first": "Xin",
"middle": [],
"last": "Suganthan",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Yao",
"suffix": ""
}
],
"year": 2006,
"venue": "Machine Learning",
"volume": "65",
"issue": "",
"pages": "247--271",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "E Ke Tang, Ponnuthurai N Suganthan, and Xin Yao. 2006. An analysis of diversity measures. Machine Learning, 65(1):247-271.",
"links": null
},
"BIBREF46": {
"ref_id": "b46",
"title": "On the likelihood that one unknown probability exceeds another in view of the evidence of two samples",
"authors": [
{
"first": "",
"middle": [],
"last": "William R Thompson",
"suffix": ""
}
],
"year": 1933,
"venue": "Biometrika",
"volume": "25",
"issue": "3/4",
"pages": "285--294",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "William R Thompson. 1933. On the likelihood that one unknown probability exceeds another in view of the evidence of two samples. Biometrika, 25(3/4):285-294.",
"links": null
},
"BIBREF47": {
"ref_id": "b47",
"title": "Introduction to the conll-2003 shared task: Language-independent named entity recognition",
"authors": [
{
"first": "Erik F Tjong Kim",
"middle": [],
"last": "Sang",
"suffix": ""
},
{
"first": "Fien",
"middle": [],
"last": "De Meulder",
"suffix": ""
}
],
"year": 2003,
"venue": "Proc. CoNLL",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Erik F Tjong Kim Sang and Fien De Meulder. 2003. Introduction to the conll-2003 shared task: Language-independent named entity recognition. In Proc. CoNLL.",
"links": null
},
"BIBREF48": {
"ref_id": "b48",
"title": "Metaphor detection with cross-lingual model transfer",
"authors": [
{
"first": "Yulia",
"middle": [],
"last": "Tsvetkov",
"suffix": ""
},
{
"first": "Leonid",
"middle": [],
"last": "Boytsov",
"suffix": ""
},
{
"first": "Anatole",
"middle": [],
"last": "Gershman",
"suffix": ""
},
{
"first": "Eric",
"middle": [],
"last": "Nyberg",
"suffix": ""
},
{
"first": "Chris",
"middle": [],
"last": "Dyer",
"suffix": ""
}
],
"year": 2014,
"venue": "Proc. ACL",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yulia Tsvetkov, Leonid Boytsov, Anatole Gershman, Eric Nyberg, and Chris Dyer. 2014. Metaphor de- tection with cross-lingual model transfer. In Proc. ACL.",
"links": null
},
"BIBREF49": {
"ref_id": "b49",
"title": "On improving the accuracy of readability classification using insights from second language acquisition",
"authors": [
{
"first": "Sowmya",
"middle": [],
"last": "Vajjala",
"suffix": ""
},
{
"first": "Detmar",
"middle": [],
"last": "Meurers",
"suffix": ""
}
],
"year": 2012,
"venue": "Proc. BEA",
"volume": "",
"issue": "",
"pages": "163--173",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sowmya Vajjala and Detmar Meurers. 2012. On im- proving the accuracy of readability classification us- ing insights from second language acquisition. In Proc. BEA, pages 163-173.",
"links": null
},
"BIBREF50": {
"ref_id": "b50",
"title": "MRC psycholinguistic database: Machine-usable dictionary, version 2.00. Behavior Research Methods, Instruments, & Computers",
"authors": [
{
"first": "Michael",
"middle": [],
"last": "Wilson",
"suffix": ""
}
],
"year": 1988,
"venue": "",
"volume": "20",
"issue": "",
"pages": "6--10",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Michael Wilson. 1988. MRC psycholinguistic database: Machine-usable dictionary, version 2.00. Behavior Research Methods, Instruments, & Com- puters, 20(1):6-10.",
"links": null
},
"BIBREF51": {
"ref_id": "b51",
"title": "Bayesian optimization of text representations",
"authors": [
{
"first": "Dani",
"middle": [],
"last": "Yogatama",
"suffix": ""
},
{
"first": "Lingpeng",
"middle": [],
"last": "Kong",
"suffix": ""
},
{
"first": "Noah A",
"middle": [],
"last": "Smith",
"suffix": ""
}
],
"year": 2015,
"venue": "Proc. EMNLP",
"volume": "",
"issue": "",
"pages": "2100--2105",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Dani Yogatama, Lingpeng Kong, and Noah A Smith. 2015. Bayesian optimization of text representations. In Proc. EMNLP, pages 2100-2105.",
"links": null
},
"BIBREF52": {
"ref_id": "b52",
"title": "Syntactic processing using the generalized perceptron and beam search",
"authors": [
{
"first": "Yue",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "Stephen",
"middle": [],
"last": "Clark",
"suffix": ""
}
],
"year": 2011,
"venue": "Computational Linguistics",
"volume": "37",
"issue": "1",
"pages": "105--151",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yue Zhang and Stephen Clark. 2011. Syntactic pro- cessing using the generalized perceptron and beam search. Computational Linguistics, 37(1):105-151.",
"links": null
}
},
"ref_entries": {
"FIGREF1": {
"uris": null,
"type_str": "figure",
"num": null,
"text": "Curriculum optimization framework."
},
"FIGREF2": {
"uris": null,
"type_str": "figure",
"num": null,
"text": "Language model score \u2022 Character language model score \u2022 Average sentence length \u2022 Verb-token ratio \u2022 Noun-token ratio \u2022 Parse tree depth \u2022 Number of noun phrases: #N P s \u2022 Number of verb phrases: #V Bs \u2022 Number of prepositional phrases: #P P s"
},
"FIGREF3": {
"uris": null,
"type_str": "figure",
"num": null,
"text": "t=1 log p(w t |w t\u2212c ..w t+c ) as follows: 5 T t=1"
},
"TABREF1": {
"text": "Training data sizes.",
"type_str": "table",
"content": "<table/>",
"num": null,
"html": null
},
"TABREF3": {
"text": "Evaluation of the impact of the curriculum of word embeddings on the downstream tasks.",
"type_str": "table",
"content": "<table><tr><td>Trimingham \" Golf \" ball .</td></tr><tr><td>Ad\u00e9lie penguin</td></tr><tr><td>\" Atriplex \" leaf UNK UNK</td></tr><tr><td>H\u1ed3ng L\u0129nh mountain</td></tr><tr><td>Anneli J\u00e4\u00e4tteenm\u00e4ki UNK cabinet</td></tr><tr><td>G\u00e4vle goat</td></tr><tr><td>Early telescope observations .</td></tr><tr><td>Scioptric ball</td></tr><tr><td>Matryoshka doll</td></tr><tr><td>Luxembourgian passport</td></tr><tr><td>Australian Cashmere goat</td></tr><tr><td>Plumbeous water redstart</td></tr><tr><td>Dageb\u00fcll lighthouse</td></tr><tr><td>Vecom FollowUs . tv</td></tr><tr><td>Syracuse Junction railroad .</td></tr><tr><td>San Clemente Island goat</td></tr><tr><td>Tychonoff plank</td></tr><tr><td>Figure 2: Most of the top lines in best-scoring</td></tr><tr><td>NER system contain named entities, although our</td></tr><tr><td>features do not annotate named entities explicitly.</td></tr></table>",
"num": null,
"html": null
},
"TABREF4": {
"text": "Parse curriculum 67.44 86.42 96.62 76.63 cbow+curric 68.26 86.49 96.48 76.54",
"type_str": "table",
"content": "<table/>",
"num": null,
"html": null
},
"TABREF5": {
"text": "Senti NER POS Parse random 63.97 82.35 96.22 69.11 curriculum 64.47 76.96 96.55 72.93",
"type_str": "table",
"content": "<table/>",
"num": null,
"html": null
}
}
}
} |