Understanding shifts in creative work will help guide AI’s impact on the media ecosystem

Enabled by artificial intelligence techniques, we are witnessing the rise of a new paradigm of computational creativity support: mixed-initiative creative interfaces put human and computer in a tight interactive loop where each suggests, produces, evaluates, modifies, and selects creative outputs in response to the other. This paradigm could broaden and amplify creative capacity for all, but has so far remained mostly confined to artificial intelligence for game content generation, and faces many unsolved interaction design challenges. This workshop therefore convenes CHI and game researchers to advance mixed-initiative approaches to creativity support.

The authors present a visual instrument developed as part of the creation of the artwork Learning to See. The artwork explores bias in artificial neural networks and provides mechanisms for the manipulation of specifically trained-for real-world representations. The exploration of these representations acts as a metaphor for the process of developing a visual understanding and/or visual vocabulary of the world. These representations can be explored and manipulated in real time, and have been produced in such a way so as to reflect specific creative perspectives that call into question the relationship between how both artificial neural networks and humans may construct meaning. Summary Memo AktenArtist, Researcher

We propose a computational framework to learn stylisation pa erns from example drawings or writings, and then generate new trajectories that possess similar stylistic qualities. We particularly focus on the generation and stylisation of trajectories that are similar to the ones that can be seen in calligraphy and gra ti art. Our system is able to extract and learn dynamic and visual qualities from a small number of user de ned examples which can be recorded with a digitiser device, such as a tablet, mouse or motion capture sensors. Our system is then able to transform new user drawn traces to be kinematically and stylistically similar to the training examples. We implement the system using a Recurrent Mixture Density Network (RMDN) combined with a representation given by the parameters of the Sigma Lognormal model, a physiologically plausible model of movement that has been shown to closely reproduce the velocity and trace of human handwriting gestures.

Recurrent Neural Networks (RNN), particularly Long Short Term Memory (LSTM) RNNs, are a popular and very successful method for learning and generating sequences. However, current generative RNN techniques do not allow real-time interactive control of the sequence generation process, thus aren't well suited for live creative expression. We propose a method of real-time continuous control and 'steering' of sequence generation using an ensemble of RNNs and dynamically altering the mixture weights of the models. We demonstrate the method using character based LSTM networks and a gestural interface allowing users to 'conduct' the generation of text. IntroductionRecurrent Neural Networks (RNN) are artificial neural networks with recurrent connections, allowing them to learn temporal regularities and model sequences. Long Short Term Memory (LSTM) [16] is a recurrent architecture that overcomes the problem of gradients exponentially vanishing [15, 1], and allows RNNs to be trained many time-steps into the past, to learn more complex programs [21]. Now, with increased compute power and large training sets, LSTMs and related architectures are proving successful not only in sequence classification [11,14,20,12], but also in sequence generation in many domains such as music [6,2,19,22], text [24,23], handwriting [10], images [13], machine translation [25], speech synthesis [28] and even choreography [4]. However, most current applications of sequence generation with RNNs is not a real-time, interactive process. Some recent implementations have used a turn-based approach, such as the online text editor Word Synth [8]. This allows a user to enter a 'seed' phrase for the RNN, 'priming' it such that the next phrase generated is conditioned on the seed. Although a very useful approach, this still does not provide real-time continuous control in the manner required for the creation of expressive interfaces. MethodAn ensemble of models is usually used to improve overall prediction accuracy. The common motivation behind this approach is that training multiple diverse models (using different architectures, parameters and/or algorithms) and then combining their predictions (through weighted or unweighted 'voting' or averaging) is likely to minimise bias and undesired variance, and thus is more likely to provide more accurate results [5]. Usually in these cases, all models are trained on the same training data.We propose a method of using an RNN ensemble, containing models trained on vastly different datasets, and dynamically altering the models' mixture weights in real-time to control the output 30th Conference on Neural Information Processing Systems (NIPS 2016),