This fascinating case is imperative reading for anyone not computer literate who wishes to understand the concept of an Artificial Neural Network.
In recent times the courts have had to grapple with the difficulties of this concept in relation to traditional computers and software. This appeal raised new questions because it involved deciding whether the use of an aspect of Artificial Intelligence, namely an Artificial Neural Network (“ANN”), in the circumstances of this case, fell under section 1(2)(c) of the Patents Act 1977 which excludes from patent protection “a program for a computer … as such”..
This issue had not yet arisen in any of the authorities before it came before Sir Anthony Mann in the Chancery Division.
The appellant company filed a patent application in 2019 for a novel technique that permits the trained ANN to align its output closer towards how a human semantically perceives content. The application was turned down because the hearing officer in the UK Intellectual Property Office found that this ANN engaged the exclusion under section 1(2)(c) and was therefore ineligible for patent protection.
In broadest outline, this ANN is trained on pairs of media files which have been tagged with text descriptions of their content (the semantic element), and a pairwise comparison of this textual, or emotive, content is used to train a separate ANN to analyse the physical properties of unknown tracks to identify their emotional and musical similarities (the property vector).
Whilst the UKIPO had accepted that this represented a significant improvement on the identified prior art, it refused the application as falling within the exclusion relating to a ‘program for a computer … as such’.
It is useful to set out in full the judge’s description of this ANN to fully understand what it involves.
“.[Let us] ..assume for the moment that the ANN itself is a hardware system (as opposed to a software emulation). A pair of music files is taken, each of which is accompanied by a natural language description of some sort of the type of music in its file in terms of how that music is perceived by a human. At its simplest the music might be described as happy, or sad, or relaxing, though the descriptions will be more complicated and wordy than that. The descriptions are in word form (hence the use of the word “semantic” and its derivatives which are used to describe this sort of feature of the music) and are to be analysed by an ANN via natural language processing software. An ANN is given instructions which enable it to assimilate the characterisation of the tracks and produce a vector or co-ordinates in a notional space (the semantic space) based on the type of music for each of the items in the pair. The similarity or difference between the semantic types of music is reflected by the distance between those two vectors (co-ordinates) in the semantic space. Two tracks of music which are semantically similar will have co-ordinates closer together; the farther apart they are in similarity the farther apart their vectors (co-ordinates) will be.”
“At the same time the same two tracks are analysed in another ANN (via parameters set by a human) for what are described as its physical properties – tone, timbre, speed, loudness and a lot of other characteristics set by the human … That analysis produces vectors (co-ordinates) in a notional “property space” (or “property embedding space” …), again with the differences or similarities in the music thus assessed reflected in the proximity of the co-ordinates. This is the ANN which will be the final operative ANN in the system.”
“The next step is a significant “trick” in the invention. The second ANN is trained to make the distances between pairs of the property co-ordinates converge or diverge in alignment with the distancing between them in the semantic space. Thus if the property space co-ordinates are farther apart than those in the semantic space, they are moved closer together, and conversely if the distancing is too close together in the property space to reflect semantic dissimilarity. This training is achieved by a process called back-propagation in which the “error” in the property space is corrected in order to make the results coincide with the training objectives. The back-propagation is done via an algorithm provided by a human and the correction is achieved by the ANN’s adjusting its own internal workings in such ways as adjusting weighting and bias in its nodes and levels of assessment. It learns from the experience without being told how to do it by a human being.” [paras 8-10]
When the above process is finished, an ANN may take any given track of music provided by a remote user, determine its physical properties and attribute a property or physical vector to it. It then ascertains music which is semantically similar by looking for tracks with physical coordinates that are close to the semantic ones and it can make a recommendation of a similar track from the data it has assessed. It does this by sending a message and a file to the remote user.
Sir Anthony Mann attached more importance than the UKIPO to the difference between hardware and software ANN. Hardware ANN is a piece of hardware which can be bought off the shelf and which contains the nodes and layers in hardware form. But an ANN can also exist in a computer emulation. In this scenario a conventional computer runs a piece of software which enables the computer to emulate the hardware ANN as if it were a hardware ANN. This is slower than a dedicated hardware ANN. However, it has the same effect.
“It should be appreciated that there is (or is said by Emotional Perception to be) a distinction between the underlying software on the computer which “creates” the emulated ANN, and the emulated ANN itself” [para 18]
The UKIPO had considered the range of UK and EPO decisions, including Aerotel Ltd v Telco Holdings Ltd and others [2007] RPC 7, and concluded that the patent application failed one of the tests for determining whether it is a computer programme, that is, whether the claimed technical effect results in the computer being made to operate in a new way. Sir Anthony found, on the contrary, that this ANN fell outside the exclusion for principal two reasons:
- He accepted the submission that an ANN is not in itself a program for a computer. He agreed with the appellant that no programmer needed to code all the detailed logical steps of the training model; the ANN adjusted itself through its training to produce a model which satisfied the training objective. Whilst an ANN may be implemented in hardware or emulated in software, the ANN lies at a higher level of generality comprising a self obtained structure, but which is not implementing code provided to it by a human. [my italics] It did not, at the level of generality at which the ANN was claimed, claim a program for a computer at all.
- In any event, the judge held that the system provided a ‘technical effect’ sufficient to avoid the exclusion. He noted that the system involved sending a file containing a recommendation to a remote user but moreover, and importantly, he held that the ANN resulting from the training process may itself constitute a technical effect
There were two fundamental questions, said the judge, that the Decision had not taken on board. 1. What is the computer, and, 2. Where is the program which is said to engage the exclusion? In most of the cases which deal with this exclusion there had been no apparent difficulty in identifying the computer and the program in issue. However, in the present case
“…the fact that a self-trained ANN is involved means that the issue is something which requires more focus than it had been given in order to resolve the questions that arise, and it transpired that asking those questions revealed significant positions on the part of UKIPO which had not hitherto clearly emerged (and indeed an arguable change of position was revealed). [para 32]
Insofar as the invention is implemented in software form, the trained ANN is not a computer, but the trained ANN is supported by a computer. In Sir Antony’s view it was “appropriate to look at the emulated ANN as, in substance, operating at a different level (albeit metaphorically) from the underlying software on the computer, and it is operating in the same way as the hardware ANN.” [para 56]. He therefore allowed the appeal and remitted the application for further processing.
This important ruling means that ANNs, which essentially create human-like user experiences through technology, and the training of these ANNs are eligible for patent protection in this country.
The post Is an Artificial Neural Network a computer programme ineligible for patent protection? appeared first on UK Human Rights Blog.