1. { M ^ { 2 } } {] timestep is 10: [-\frac { M ^ { 2 } } { 4] timestep is 11: [-\frac { M ^ { 2 } } { 4 }] timestep is 12: [-\frac { M ^ { 2 } } { 4 } \tan] timestep is 13: [-\frac { M ^ { 2 } } { 4 } \tan (] timestep is 14: [-\frac { M ^ { 2 } } { 4 } \tan ( \frac] timestep is;{ M ^ { 2 } } { 4 } \tan;\frac { M ^ { 2 } } { 4 } \tan ( \frac { p \pi } { 2] timestep is 21: [-\frac { M ^ { 2 } } { 4 } \tan ( \frac { p \pi } { 2 }] timestep is
2. Syntax-Directed Recognition of Hand-Printed Two-Dimensional Mathematics;Robert H Anderson;Symposium on Interactive Systems for Experimental Applied Mathematics: Proceedings of the
3. Learning long-term dependencies with gradient descent is difficult;Y Bengio;IEEE Transactions on Neural Networks,1994
4. Long Short-Term Memory;Sepp Hochreiter;Neural Computation,1997