Explore more searches like Cadrec Roseau |
- 1471×580
- 1117×1117
- 2048×1536
- 780×1024
- 1000×1498
- 1224×1632
+or+debiasing+methods+has+attained+remarkable+gains.+However%252C+they+still+suffer+from+(1)+over-smoothing+node+embeddings+caused+by+recursive+convolutions+with+GNNs%252C+and+(2)+the+skewed+distribution+of+interactions+due+to+popularity+and+user-individual+biases.+This+paper+proposes+a+contextualized+and+debiased+recommender+model+(CaDRec).+To+overcome+the+over-smoothing+issue%252C+we+explore+a+novel+hypergraph+convolution+operator+that+can+select+effective+neighbors+during+convolution+by+introducing+both+structural+context+and+sequential+context.+To+tackle+the+skewed+distribution%252C+we+propose+two+strategies+for+disentangling+interactions%3a+(1)+modeling+individual+biases+to+learn+unbiased+item+embeddings%252C+and+(2)+incorporating+item+popularity+with+positional+encoding.+Moreover%252C+we+mathematically+show+that+the+imbalance+of+the+gradients+to+update+item+embeddings+exacerbates+the+popularity+bias%252C+thus+adopting+regularization+and+weighting+schemes+as+solutions.+Extensive+experiments+on+four+datasets+demonstrate+the+superiority+of+the+CaDRec+against+state-of-the-art+(SOTA)+methods.+Our+source+code+and+data+are+released+at+https%3a%252F%252Fgithub.com%252FWangXFng%252FCaDRec.&cdnurl=https%3a%2f%2fth.bing.com%2fth%2fid%2fR.c62b795420366eaca80f65e1bf3f55a5%3frik%3dzVfHoG8vncx9Mg%26pid%3dImgRaw%26r%3d0&exph=630&expw=1200&q=Cadrec+Roseau&simid=608051603618731387&FORM=IRPRST&ck=6BBBB346660482D8A7DFD47B1D4401AA&selectedIndex=20&itb=0)
- 1122×631
Some results have been hidden because they may be inaccessible to you.Show inaccessible results
Explore more searches like Cadrec Roseau |
Feedback
