[86] E. Trentin, L. Lusnig, and F. Cavalli. Parzen neural networks: Fundamentals, properties, and an application to forensic anthropology. *Neural Networks* 97: 137-151 (2018).

[85] M. Bongini, L. Rigutini, and E. Trentin. Recursive Neural Networks for Density Estimation Over Generalized Random Graphs. *IEEE Trans. on Neural Networks and Learning Systems* 29(11): 5441-5458 (2018).

[84] E. Trentin and E. Di Iorio. Nonparametric small random networks for graph-structured pattern recognition. *Neurocomputing* 313: 14-24 (2018).

[83] E. Trentin. Soft-Constrained Neural Networks for Nonparametric Density Estimation. *Neural Processing Letters* 48(2): 915-932 (2018).

[82] M. Bongini, A. Freno, V. Laveglia, and E. Trentin. Dynamic Hybrid Random Fields for the Probabilistic Graphical Modeling of Sequential Data: Definitions, Algorithms, and an Application to Bioinformatics. *Neural Processing Letters* 48(2): 733-768 (2018).

[81] E. Trentin, F. Schwenker, N. El Gayar, and H. M. Abbas. Off the Mainstream: Advances in Neural Networks and Machine Learning for Pattern Recognition. *Neural Processing Letters* 48(2): 643-648 (2018).

[80] V. Laveglia and Edmondo Trentin. A Refinement Algorithm for Deep Learning via Error-Driven Propagation of Target Outputs. * Proc. of ANNPR 2018*: 78-89, Springer, 2018.

[79] E. Trentin. Maximum-Likelihood Estimation of Neural Mixture Densities: Model, Algorithm, and Preliminary Experimental Evaluation. * Proc. of ANNPR 2018*: 178-189, Springer, 2018.

[78] L. Pancioni, F. Schwenker, and Edmondo Trentin. *Artificial Neural Networks in Pattern Recognition - 8th IAPR TC3 Workshop, ANNPR 2018, Siena, Italy, September 19-21, 2018, Proceedings*. Lecture Notes in Computer Science 11081, Springer 2018, ISBN 978-3-319-99977-7.

[77] F. Cavalli, L. Lusnig, and E. Trentin. Use of Pattern Recognition and Neural Networks for Non-Metric Sex Diagnosis From Lateral Shape of Calvarium: An Innovative Model for Computer-Aided Diagnosis in Forensic and Physical Anthropology. *International Journal of Legal Medicine* 131(3): 823-833 (2017) doi:10.1007/s00414-016-1439-8.

[76] E. Trentin. Soft-Constrained Nonparametric Density Estimation with Artificial Neural Networks. In *Proc. of ANNPR 2016*: 68-79, Springer (2016).

[75] M. Bongini, V. Laveglia, and E. Trentin. A Hybrid Recurrent Neural Network/Dynamic Probabilistic Graphical Model Predictor of the Disulfide Bonding State of Cysteines from the Primary Structure of Proteins. In *Proc. of ANNPR 2016*: 257-268, Springer, 2016.

[74] F. Schwenker, H. M. Abbas, N. El Gayar, and E. Trentin (Eds.). *Artificial Neural Networks in Pattern Recognition - 7th IAPR TC3 Workshop, ANNPR 2016, Ulm, Germany, September 28-30, 2016, Proceedings.* Lecture Notes in Computer Science 9896, Springer (2016).

[73] E. Trentin and M. Bongini. Probabilistically Grounded Unsupervised Training of Neural Networks. In M.E. Celebi and K. Aydin (Eds.) *Unsupervised Learning Algorithms*: 533-558 Springer (2016).

[72] E. Trentin. Maximum-likelihood normalization of features increases the robustness of neural-based spoken human-computer interaction. *Pattern Recognition Letters* 66(C): 71--80 (2015).

[71] E. Trentin, S. Scherer, F. Schwenker. Emotion recognition from speech signals via a probabilistic echo-state network. *Pattern Recognition Letters* 66(C): 4-12 (2015).

[70] M. Aste, M. Boninsegna, A. Freno, E. Trentin. Techniques for dealing with incomplete data: a tutorial and survey. *Pattern Analysis and Applications* 18(1): 1-29 (2015).

[69] M. Bongini, F. Schwenker, E. Trentin. On Semi-Supervised Clustering. In Celebi, M.E. (Ed.) *Partitional Clustering Algorithms*: 277-311 Springer (2015).

[68] E. Trentin. Libero arbitrio e macchine intelligenti. In Tugnoli, C. (Ed.) *Libero arbitrio - Teorie e prassi della libertà*: Liguori, 2014. ISBN: 978-88-207-5330-6. In Italian.

[67] F. Schwenker and E. Trentin. Pattern Classification and Clustering: a
Review of Partially Supervised Learning Approaches. *Pattern Recognition Letters*
37: 4-14 (2014).

[66] F. Schwenker and E. Trentin. Partially Supervised Learning for Pattern
Recognition. *Pattern Recognition Letters* 37: 1-3 (2014).

[65] I. Castelli and E. Trentin. Combination of supervised and unsupervised
learning for training the activation functions of neural
networks. *Pattern Recognition Letters* 37: 178-191 (2014).

[64] M. Glodek, E. Trentin, F. Schwenker, and G. Palm. Hidden Markov Models
With Graph Densities for Action Recognition. In *Proc. of IJCNN 2013 (INNS
IEEE International Joint Conference on Neural Networks)*, Dallas (2013).

[63] M. Bongini and E. Trentin. Towards a Novel Probabilistic Graphical Model
of Sequential Data: A Solution to the Problem of Structure Learning and an
Empirical Evaluation. In *Proc. of ANNPR 2012*: 82-92 (2012).

[62] E. Trentin and M. Bongini. Towards a Novel Probabilistic Graphical Model
of Sequential Data: Fundamental Notions and a Solution to the Problem of
Parameter Learning. In *Proc. of ANNPR 2012*: 72-81 (2012).

[61] N. Mana, F. Schwenker, and E. Trentin (Eds.). *Artificial Neural Networks in Pattern Recognition - 5th INNS IAPR TC 3 GIRPR Workshop, ANNPR 2012, Trento, Italy, September 17-19, 2012. Proceedings.* Lecture Notes in Computer Science 7477, Springer 2012, ISBN 978-3-642-33211-1.

[60] F. Schwenker and E. Trentin (Eds.). *Partially Supervised Learning -
First IAPR TC3 Workshop, PSL 2011, Ulm, Germany, September 15-16, 2011,
Revised Selected Papers..* Lecture Notes in Computer Science 7081, Springer
2012, ISBN 978-3-642-28257-7.

[59] A. Freno and E. Trentin. *Hybrid Random Fields. A Scalable Approach
to Structure and Parameter Learning in Probabilistic Graphical Models.*
Springer, ISBN 978-3-642-20307-7, 2011.

[58] I. Castelli and E. Trentin. Semi-Unsupervised Weighted
Maximum-Likelihood Estimation of Joint Densities for the Co-Training of Adaptive Activation Functions. In *Proceedings of the 1st IAPR-TC3 Workshop on
Partially Supervised Learning (PSL 2011)*, Ulm (Germany), 2011.

[57] I. Castelli and E. Trentin. Supervised and Unsupervised Co-Training of
Adaptive Activation Functions in Neural Nets.
In *Proceedings of the 1st IAPR-TC3 Workshop on
Partially Supervised Learning (PSL 2011)*, Ulm (Germany), 2011.

[56] E. Trentin, L. Lusnig and F. Cavalli. Comparison of Combined Probabilistic Connectionist Models
in a Forensic Application. In *Proceedings of the 1st IAPR-TC3 Workshop on
Partially Supervised Learning (PSL 2011)*, Ulm (Germany), 2011 (in press).

[55] A. Freno, E. Trentin and M. Gori. Kernel-Based Hybrid Random Fields
for Nonparametric Density Estimation. In *Proceedings of ECAI 2010*,
pages 427-432.

[54] E. Trentin, S. Scherer and F. Schwenker. Maximum Echo-State-Likelihood
Networks for Emotion Recognition. In *Proceedings of ANNPR 2010 (Artificial
Neural Networks in Pattern Recognition, Fourth IAPR Workshop)*, pages 60-71,
Cairo
(Egypt), April 2010.

[53] E. Trentin, S. Zhang and M. Hagenbuchner. Recognition of Sequences of Graphical Patterns. In *Proceedings of ANNPR 2010 (Artificial
Neural Networks in Pattern Recognition, Fourth IAPR Workshop)*, pages 48-59, Cairo
(Egypt), April 2010.

[52] E. Trentin and A. Freno. Probabilistic interpretation of neural
networks for statistical and sequential pattern recognition. In *Innovations in Neural Information
Paradigms and Applications*, Springer, SCI 247-0155, pages 155-182 (2009).

[51] E. Trentin and E. Di Iorio. Classification of graphical data made
easy. *Neurocomputing* 73 (1-3): 204-212 (2009).

[50] S. Scherer, E. Trentin, F. Schwenker, and G. Palm. Approaching emotion
in human computer interaction. In *Proceedings of the International Workshop on
Spoken Dialogue Systems*, (satellite event of the IEEE ASRU 2009), Kloster Irsee (Germany), Dec. 2009.

[49] A. Freno, E. Trentin, and M. Gori. A hybrid random field model for scalable statistical learning. *Neural Networks* 22(5-6): 603-613 (2009).

[48] E. Trentin and L. Rigutini. A Maximum-Likelihood Connectionist Model
for Unsupervised Learning over Graphical Domains. In *Proceedings of ICANN
2009*, (1) 40-49.

[47] A. Freno, E. Trentin, and M. Gori. Scalable Pseudo-Likelihood Estimation in Hybrid Random Fields. In J.F. Elder, F. Fogelman-Soulie', P. Flach, and M. Zaki (eds.), *Proceedings of the 15th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2009)*. ACM, 2009, 319--327.

[46] E. Trentin and A. Freno. Unsupervised Nonparametric Density Estimation: A Neural Network Approach. In *Proceedings of the International Joint Conference on Neural Networks (IJCNN 2009)*. Atlanta, 2009, 3140--3147.

[45] A. Freno, E. Trentin, and M. Gori. Scalable Statistical Learning: A Modular Bayesian/Markov Network Approach. In *Proceedings of the International Joint Conference on Neural Networks (IJCNN 2009)*. Atlanta, 2009, 890--897.

[44] E. Trentin and E. Di Iorio. Classification of molecular structures
made easy. *Proceedings of IJCNN 2008*,
Hong Kong, June 2008, 3241-3246.

[43] E. Trentin and E. Di Iorio. Unbiased SVM Density Estimation with Application to Graphical Pattern Recognition. In *Proceedings of ICANN 2007*,
Porto, September 2007, (2) 271-280.

[42] E. Trentin and E. Di Iorio. A Simple and Effective Neural Model for
the Classification of Structured Patterns. In *Proceedings of KES 2007*,
Vietri Sul Mare (Italy), September 2007, (1) 9-16.

[41] V. Montagnani and E. Trentin. Hidden Markov Models and Multiple
Transfer Functions for Voice Subtraction in an Acoustic
Dosimeter. *Circuits, Systems & Signal Processing*, 26(3):311-323,
June 2007.

[40] P. Fiengo, G. Giambene and E. Trentin. Neural-based downlink scheduling algorithm for broadband wireless networks.
*Computer Communications*, 30(2):207--218, January 2007.

[39] E. Trentin and M. Gori. Inversion-based nonlinear adaptation of noisy
acoustic parameters for a neural/HMM speech recognizer. *Neurocomputing*, 70(1-3):398--408, 2006.

[38] E. Trentin. A Novel Connectionist-Oriented Feature Normalization
Technique. In *Proceedings of ICANN 2006 (International Conference on
Artificial Neural Networks)*, 410-416, Athena, Greece, September 2006.

[37] E. Trentin. Simple and Effective Connectionist Nonparametric Estimation of Probability Density Functions.
In *Proceedings of ANNPR 2006 (Artificial Neural Networks in Pattern Recognition, Second IAPR Workshop)*: 1-10, Ulm, Germany, August
31-September 2, 2006.

[36] E. Trentin and M. Gori. Feature normalization via ANN/HMM inversion for speech recognition under noisy conditions.
In *Proceedings of MMSP 2005, IEEE International Workshop on
Multimedia Signal Processing*, Shanghai, China, Oct-Nov 2005.

[35] E. Trentin and M. Gori. Robust Combination of Neural Networks and
Hidden Markov Models for Speech Recognition. *IEEE Transactions on Neural Networks*, 14(6):1519--1531, November 2003.

[34] E. Trentin and M. Matassoni. Noise-tolerant speech recognition: the
SNN-TA approach. *Information Sciences* (Special Issue on Intelligent Systems for Speech and Language), 156(1-2):55--69, November 2003.

[33] E. Trentin. Nonparametric Hidden Markov Models: Principles and
Applications to Speech Recognition. In B. Apolloni, M. Marinaro and
R. Tagliaferri (Editor), *Neural Nets - WIRN
Vietri - 2003*, 3--24. Berlin, 2003, Springer-Verlag.

[32] E. Trentin, M. Matassoni and M. Gori.
Evaluation on the Aurora 2 Database of Acoustic Models that are
less Noise-sensitive.
In *Proceedings of Eurospeech 2003*, Geneva, Switzerland,
September 2003.

[31] E. Trentin, L. Magnoni and A. Andronico. Toward A Modular
Connectionist Model of Local Chlorophyll Concentration from Satellite
Images. In *Proceedings of IJCNN03, IEEE-INNS International Joint
Conference on Neural Networks*, Portland (Oregon), July 2003.

[30] E. Trentin, F. Brugnara, Y. Bengio, C. Furlanello and R. De Mori.
Statistical and neural network models for speech recognition. In R. Daniloff
(Ed.) *Connectionist accounts of clinical and normal language*, 213--264, Mahwah, New Jersey, 2002. Lawrence Erlbaum Associates.

[29] E. Trentin and M. Gori. Toward Noise-tolerant Acoustic Models. In *Proceedings of Eurospeech2001- Scandinavia*, Aalborg, Denmark, September 2001.

[28] E. Trentin and M. Gori. Continuous Speech Recognition with a Robust
Connectionist/Markovian Hybrid Model. In *Proceedings of ICANN2001*, Vienna, Austria, August 2001. *Winner of the Best Paper Award*.

[27] E. Trentin and D. Giuliani. A mixture of recurrent neural networks
for speaker normalization. *Neural Computing & Applications*,
July 2001, 10:120-135.

[26] E. Trentin. Networks with trainable amplitude of activation
functions. *Neural Networks* 14 (July 2001), 471-493.

[25] E. Trentin and M. Gori. A survey of hybrid ANN/HMM models for automatic
speech recognition. *Neurocomputing*, 37(1/4); 91-126, March 2001.

[24] E. Trentin. Robust Combination of Neural Networks and Hidden Markov Models for Speech Recognition. PhD Thesis, University of Florence (Italy), February 2001.

[23] E. Trentin and M. Matassoni. The regularized SNN-TA model for recognition
of noisy speech. In *Proceedings of IJCNN 2000 (International Joint Conference
on Neural Networks)*, Como (Italy), 24-27 July 2000.

[22] E. Trentin and R. Cattoni. Learning perception for indoor robot
navigation with a hybrid HMM/Recurrent neural networks approach. *Connection
Science*, 11(3/4): 243-265, December 1999. Special Issue on Adaptive
Robots.

[21] M. Matassoni, M. Omologo, L. Cristoforetti, D. Giuliani, P. Svaizer,
E. Trentin, and E. Zovato. Some results on the development of a hands-free
speech recognizer for car-environment. In *Proceedings of the 1999 international
workshop on Automatic Speech Recognition and Understanding (ASRU)*,
Keystone, Colorado, USA, December 12-15 1999.

[20] E. Trentin. Activation functions with learnable amplitude. In *Proceedings
of IJCNN99, International Joint Conference on Neural Networks*, Washington,
DC, July 1999.

[19] E. Trentin and M. Gori. A tutorial on connectionist and hybrid
HMM/connectionist systems for speech recognition. In N. Kasabov, editor,
*IJCNN99 Tutorials Track 8 (Book on CD-ROM, Chapter 3): Speech and Language
Processing*, Dunedin, New Zealand, 1999. University of Otago.

[18] E. Trentin and M. Matassoni. Robust segmental-connectionist learning
for recognition of noisy speech. In *Proceedings of the first Workshop
on Robust Methods for Speech Recognition in Adverse Conditions*, pages
159 - 162, Tampere, Finland, May 25-26 1999.

[17] C. Furlanello, D. Giuliani, S. Merler and E. Trentin. Model selection
of combined neural nets for speech recognition. In A. J. C. Sharkey, editor,
*Combining Artificial Neural Nets - Ensemble and Modular Multi-Net Systems*,
pages 205 - 233, London, UK, 1999. Springer-Verlag.

[16] E. Trentin. HMMs for acoustic modeling: Beyond the problem of local
stationarity. In *Proceedings of IX AIA Conference on Computational
Aspects of Phonetics*, volume XXVI, Venezia, 1999.

[15] E. Trentin and M. Gori. Combining neural networks and hidden Markov
models for speech recognition. In M. Marinaro and R. Tagliaferri, editors,
*Neural Nets - WIRN Vietri - 98*, pages 63 - 79, Berlin, Germany,
1999. Springer-Verlag.

[14] E. Trentin. Learning the amplitude of activation functions in layered
networks. In M. Marinaro and R. Tagliaferri, editors, *Neural Nets -
WIRN Vietri - 98*, pages 138 - 144, Berlin, Germany, 1999. Springer-Verlag.

[13] E. Trentin and R. Cattoni. A hybrid framework for indoor robot
navigation. In M. Marinaro and R. Tagliaferri, editors, *Neural Nets
- WIRN Vietri - 98*, pages 255 - 263, Berlin, Germany, 1999. Springer-Verlag.

[12] E. Trentin, Y. Bengio, C. Furlanello and R. De Mori. Neural networks
for speech recognition. In R. De Mori, editor, *Spoken Dialogues with
Computers*, pages 311-361, London, UK, 1998. Academic Press.

[11] M. Boninsegna, T. Coianiz and E. Trentin. Estimating the crowding
level with a neuro-fuzzy classifier. *Journal of Electronic Imaging*,
6(3):319-328, July 1997.

[10] E. Trentin and D. Giuliani. Speaker normalization with a mixture
of recurrent networks. In *Proceedings of ESANN97, European Symposium
on Artificial Neural Networks*, Bruges, Belgium, April 1997.

[9] E. Trentin and R. Cattoni. A hybrid HMM/recurrent neural networks
approach to indoor robot navigation. In *Proceedings of RWC97 - Real
World Computing Symposium*, Tokyo, Japan, January 1997.

[8] C. Furlanello, D. Giuliani, E. Trentin and S. Merler. Speaker normalization
and model selection of combined neural nets. *Connection Science*,
9(1):31-50, January 1997. Special Issue on Combining Neural Nets.

[7] E. Trentin, D. Giuliani and C. Furlanello. Spectral mapping: A comparison
of connectionist approaches. In M. Marinaro and R. Tagliaferri, editors,
*Neural Nets. WIRN Vietri-96*, pages 270-277, Berlin, May 1996. Springer
- Verlag.

[6] E. Trentin, C. Furlanello and D. Falavigna. An evaluation criterion
for connectionist systems in classification tasks. In *Proceedings of
First Workshop on Evaluation criteria of neural nets efficiency in industrial
applications*, IIASS, Vietri sul Mare (SA), December 1995.

[5] C. Furlanello, D. Giuliani, E. Trentin and D. Falavigna. Applications
of generalized radial basis functions in speaker normalization and identification.
In *Proceedings of ISCAS '95, IEEE International Symposium on Circuit
and Systems*, pages 867-874, Seattle WA, April, 30 - May, 3 1995.

[4] C. Furlanello, D. Giuliani and E. Trentin. Connectionist speaker
normalization with Generalized Resource Allocating Networks. In G. Tesauro,
D. S. Touretzky and T. K. Leen, editors, *Advances in Neural Information
Processing Systems 7*, pages 1704-1707, Cambridge MA, 1995. MIT Press.

[3] E. Trentin and P. Miotti. HE: una macchina astratta per la gestione
di sistemi ipermediali. *Note di Software*, 51:35-43, March 1991.
In Italian.

[2] E. Trentin. I frattali - parte II. *Bit*, 107:126-131, July
- August 1989. In Italian.

[1] E. Trentin. I frattali - parte I. *Bit*, 106:97-101, June 1989.
In Italian.