Báo cáo khoa học: "LEARNING TO RESOLVE BRIDGING REFERENCES" ppt

... /1We will use the term bridging descriptions to indicate bridging references realized by deﬁnite descriptions, equatedhere with noun phrases with determiner the, like the top.2Following (Poesio ... of bridging referencesproposed by Vieira and Poesio (2000). ‘Mereological’ bridging references are one of the the ‘WordNet’ bridging classes, whichcover cases where the information required to ... due in part to missing entriesand / or relations; in part to the fact that because ofthe monotonic organization of information in Word-Net, complex searches are required even to ﬁnd ap-parently...

Tài liệu Báo cáo khoa học: "Learning to Translate with Multiple Objectives" doc

... between pareto vs. non-paretopoints ignores the fact that 2nd-place non-paretopoints may also lead to good practical solutions. Abetter approach may be to adopt a graded deﬁnitionof Pareto optimality ... weight vector w, takes in a foreign sen-tence f and returns a translated hypothesis h. Theargmax operates in vector space and our goal is to ﬁnd w leading to hypotheses on the Pareto Frontier.In ... solutions to Eq. 2 also solutions to Eq. 1 (i.e. pareto-optimal)and vice-versa? The theory says:Theorem 1. Sufﬁcient Condition: If w∗is solution to Eq. 2, then it is weakly pareto-optimal....

Tài liệu Báo cáo khoa học: "Learning to Find Translations and Transliterations on the Web" doc

... Thus, we also associate features to each Chinese token (characters or words) to reflect the likelihood of the token being part of the translation. We describe how to train a CRF model for identifying ... potentially be used to assist translators to find the most common translation for a given term, or to supplement a bilingual terminology bank (e.g., adding multilingual titles to existing Wikipedia); ... source term and partial translation to a search engine is a good strategy used by many translators. Unfortunately, the user still has to sift through snippets to find the translations. For a given...

Tài liệu Báo cáo khoa học: "Learning to Follow Navigational Directions" pdf

... itis her task to describe the path to the instructionfollower, who cannot see the reference path. Oursystem learns to interpret these navigational direc-tions, without access to explicit linguistic ... θTφ(s, a) most closely approximatesE[R(s, a)]. To learn these weights θ we useSARSA (Sutton and Barto, 1998), an online learn-ing algorithm similar to Q-learning (Watkins andDayan, 1992).Algorithm ... spatial word. The toprow shows the weights of allocentric (landmark-centered) features. For example, the top left ﬁgure showsthat when the word above occurs, our policy prefers to go to the north...

Tài liệu Báo cáo khoa học: "Learning to Compose Effective Strategies from a Library of Dialogue Components" doc

... that the key to maximum dialogue ef-fectiveness is to listen to the user. This paper de-scribes the development of an adaptive dialogue sys-tem that uses the feedback of users to automaticallyimprove ... harder to detect than others, thesystem’s classiﬁcation accuracy depends largely onthe user’s barrier. To prevent classiﬁcation accuracydistorting the user evaluations, we asked the subjects to ... MDP(see Sutton and Barto, 1998). We applied the algo-rithm of (Singh et al., 2002), as their experimentalset-up is similar to ours, constisting of: generationof (limited) exploratory dialogue...

Tài liệu Báo cáo khoa học: "Learning to Recognize Tables in Free Text" pptx

... 12: 13: Net tons Capability 14: produced utilization 15: Week to March 14 1,633,000 75.8~ 16: Week to March 7 1,570,000 71.9~ 17: Year to date 15,029,000 66.9~ 18: Year earlier to date 18,431,000 ... Learning to Recognize Tables in Free Text Hwee Tou Ng Chung Yong Lim Jessica Li Teng Koo DSO National Laboratories 20 Science Park Drive, Singapore 118230 {nhweetou, ichungyo, kliteng}@dso, ... Fared Stock's 'Bid'* Initial Date** Reaction*** Bidder (Target Company) TWAICarl Ic~h- (USAir Group) $52 +5 3/8 to 49 1/8 3/4/87 Outcome Bid, seen a ploy to get USAir to buy...

Tài liệu Báo cáo khoa học: "Learning to Behave by Reading" pot

...

Báo cáo khoa học: "Learning to Translate with Source and Target Syntax" pot

... 47target glue 36 19match 11 29mismatch 53 52Table 4: Moving from string -to- string (s -to- s) ex-traction to fuzzy tree -to- tree (t -to- t) extraction de-creases glue rule usage and increases the frequencyof ... sam-pling to sum over equivalent derivations during de-coding, and the latter uses dynamic programminghuman automaticstring -to- string 198,445 142,820max nested 78,361 64,578tree -to- string ... count is the to- tal count of all the rules. This means that there is aone -to- one correspondence between the rules ex-tracted by fuzzy tree -to- tree extraction and hierar-chical string -to- string...

Báo cáo khoa học: "Learning to Win by Reading Manuals in a Monte-Carlo Framework" pot

... approach for lever-aging automatically extracted textual knowl-edge to improve the performance of controlapplications such as games. Our ultimate goalis to enrich a stochastic player with high-level ... text to better select actions.The key technical challenge in leveraging textualknowledge is to automatically extract relevant infor-mation from text and incorporate it effectively into acontrol ... it is impractical to manually anno-tate the information that would be relevant to thosestates. Instead, we propose to learn text analysisbased on a feedback signal inherent to the controlapplication,...

Báo cáo khoa học: "Learning to Grade Short Answer Questions using Semantic Similarity Measures and Dependency Graph Alignments" ppt

... features, we use Support Vec-tor Machines (SVM) to produce a combined real-number grade. Finally, we build an Isotonic Regres-sion (IR) model to transform our output scores ontothe original [0,5] ... fo-cus of this work.An automatic short answer grading system is onethat automatically assigns a grade to an answer pro-vided by a student, usually by comparing it to oneor more correct answers. ... community.Speciﬁcally, we seek answers to the followingquestions. First, to what extent can machine learn-ing be leveraged to improve upon existing ap-proaches to short answer grading. Second, does...

Xem thêm