Belgian Bacon Pudding. A recipe from IBM’s Watson SuperComputer.

At a recent event with IBM at The Griffin, the Incogvino JHB team got to experience some cognitive cooking (and human wine pairings). The recipe below was generated by Watson, IBM’s SuperComputer.

Inputs used to generate this recipe:

• Primary ingredient: Bacon
• Primary style: Belgian
• Unique recipe, never seen before.

Pairs with: We don’t know! The Chef changed this course because he just didn’t “get” the flavours. Thankfully we forced him to bring out his trial dishes. Tres #YUM! What pairs with bacon?

Belgian bacon and porcini mushroom pudding

500ml cream

110g bacon: rendered, fat reserved

porcini mushroom powder: 2 teaspoons

ground black pepper: 1/4 teaspoons

granulated sugar: 3/4 cup

egg yolks: 2

sheet gelatin : 4 pieces, in cold water

500ml buttermilk

unsalted butter: 7 tablespoons

almond flour : 1/2 cup, finely ground

all-purpose flour: 1/2 cup

fine sea salt: pinch, more as needed

Icing sugar,: 1 1/4 cup, more as needed

large egg whites: 4

extra virgin olive oil : 2 tablespoons

orange: juice and zest of one

honey: 1 tablespoon

ground cumin : pinch

ground caraway: pinch

golden raisins: 1/2 cup

dried figs: 1/2 cup, stem tips removed and roughly chopped

Method

Make the bacon-porcini pudding, gently heat the cream to approximately 65˚C; add the warm rendered bacon and fat and infuse for at least 4-6 hours, chilled. After infusing, strain the cream and discard the bacon and congealed fat solids. Add the mushroom powder and black pepper and bring to a boil. Meanwhile, whisk together the sucrose and egg yolks. Temper the hot cream into the yolk mixture and return to low heat, cooking just to 85˚C Remove from heat and add the gelatin. Temper the mixture into the buttermilk and blend well with an immersion blender. Divide into glasses or serving dishes and chill to set.

Next, prepare the walnut financier. Gently cook the butter to a light brown color and reserve warm. Combine the dry ingredients; in another large mixing bowl, manually whip egg whites just until frothy and yellow color dissipates. Whisk in the flour mixture

Slowly whisk in the warm butter, followed by the olive oil, ensuring complete emulsification;chill. Divide the mixture into silicon baking molds; bake at 150˚C until lightly browned and cooked through. Cool and break the financier into small pieces.

To prepare the spiced fruit compote, combine the orange juice and zest, honey and spices and gently warm. Pour the orange-honey mixture over the raisins and figs and macerate several hours, chilled.

To assemble, divide the dried fruit mixture among each set pudding and top with the torn financier pieces. Garnish with a few grains of Maldon salt and a dusting of icing sugar.

The post Belgian Bacon & Mushroom Pudding. A recipe from IBM’s Watson #CognitiveCooking appeared first on Incogvino.

Inputs used to generate this recipe:

• Primary ingredient: Lamb
• Primary style: Traditional
• Combination recipe using flavour profiling (not necessarily unique).

Pairs with: Rust en Vrede Cabernet Sauvignon 2012.

Deboned loin of lamb with pinenut crust, tarragon gnocchi, savoy cabbage & anchovy stuffing. A recipe from IBM’s Watson.

Lamb

1.15kg toasted pine kernels

4kg butter, softened

10 clove garlic

100 g thyme, leaves only

200 g leaves parsley, chopped

1kg freshwhite breadcrumbs

pinches nutmeg

pinches chinese five spice

olive oil, for frying

20 x 200 g portions lamb loin

Preparation Method:

Put the pine nuts, butter, garlic, herbs, breadcrumbs and spices into a food processor and blend to a coarse paste.

Roll out the pine nut paste between 2 sheets of greaseproof paper until large enough to cover the portions of lamb, then cut in half and chill in the fridge.

Heat a little olive oil in a frying pan, and fry the lamb portions until cooked to your liking. Rest the meat for 10 minutes. 

Preheat the grill. Lay a portion of the pine nut crust on top of each lamb portion and cook under the grill until the crust is golden brown.

Tarragon gnocchi

9kg potatoes, peeled

salt and pepper

10 tbsp chopped tarragon

10 tbsp chopped flatleaf parsley

75g plain flour

glug of olive oil

salt and pepper

Preparation method:

To make the gnocchi, cook the potatoes in a pan of as little water as possible until tender. Drain and mash or rice the potatoes while they are still warm.

Turn out the mash onto a work surface, season with salt and pepper and add the herbs. Slowly add the flour in batches, kneading gently by folding the edges of the dough into the center and pressing down lightly before sprinkling in each further addition of flour, until the dough is soft. Don’t overwork the dough or you will spoil the structure and the gnocchi will be powdery on the outside and pulpy on the inside.

Roll the dough into a Frankfurter-sized sausage. Cut the sausage into 3cm pieces.

Press each piece of dough against the back of a fork so that it curves slightly and the prongs leave indentations (which will help the gnocchi hold the sauce).

Bring a large pan of salted water to the boil. Add the olive oil and drop in the gnocchi. Cook for a minute, or until the gnocchi rises to the surface. Remove from the pan, plunge into a bowl of ice-cold water and drain.

Caponata

9 medium aubergines, chopped

salt and freshly ground black pepper

6 tbsp extra virgin olive oil, plus extra for drizzling

2 onion, chopped

4 celery sticks, chopped

800g chopped tomatoes

200g green olives, pitted and sliced

6 tbsp capers, drained and chopped

60ml red wine vinegar

3 tbsp sugar, or to taste

handful flatleaf parsley, chopped

Preparation method:

Sprinkle the aubergines with salt and leave to drain in a colander for 30 minutes.

Heat some of the olive oil in a saucepan and brown the aubergine on a moderate heat for 10 minutes. When cooked, set aside and allow to cool to room temperature.

In a separate saucepan, heat the remaining olive oil and sauté the onion along with the celery and tomatoes. Season with salt and freshly ground black pepper. Add the olives and cook for 20 minutes.

Add the cooled aubergine and the capers.

In a separate bowl, mix together the red wine vinegar and sugar. Add this to the pan and cook for 10 minutes. It is ready when the red wine vinegar has been absorbed & add parsley

Mint gel

¼ cup Fresh mint

1/3 cup Simple Syrup

1/3 cup Water

1 t Sugar

½ t Agar

Preparation Method:

Blanch the mint, shock n dry, Blend syrup with water and mint

Place in bowl in ice bath for 20 minutes. Then strain.

Combine sugar and agar.

Place the mint liquid into a blender on low.

Sift in the agar sugar mixture and blend for 1 minute.

Pour into a sauce pan and bring to boil, simmering for 5 minutes.

Pour into dish and cover chill for 10 minutes.

Blend the mixture till smooth, stain and store.

The post Deboned Loin of Lamb with Pinenut Crust. A recipe from IBM’s Watson #CognitiveCooking appeared first on Incogvino.

Swiss Thai Quiche, a recipe from IBM’s Watson SuperComputer

At a recent event with IBM at The Griffin, the Incogvino JHB team got to experience some cognitive cooking (and human wine pairings). The recipe below was generated by Watson, IBM’s SuperComputer.

Inputs used to generate this recipe:

• Primary ingredient: Asparagus
• Primary style: Swiss & Thai
• Unique recipe, never seen before.

Pairs with: Jordan Unoaked Chardonnay 2014

Swiss Thai Quiche. A recipe from IBM’s Watson.

155g Butter: divided

82ml Water: divided

2 Egg Yolk: divided

1.5 tsp salt: divided

225g plain flour: sifted

115g Lemongrass: peeled, tender white part only, thinly sliced

115g Leeks:, white part only

25 Asparagus tips: SHORT ONE

3 Eggs

225ml Heavy Cream

225 ml Plain whole-milk yogurt

Mild Curry Powder: 3/8 tsp

Method

Make the pastry dough, place the flour in the bowl of an electric mixer fit with the paddle attachment. Add 1 tsp of salt, 1 egg yolk, and 2.5 oz of water, and mix over low speed. Make sure your butter is room-temperature, and add it, diced, continuing to mix until homogeneous. Take the dough out of the bowl and knead by hand for 1 minute. Shape into a ball, cover with plastic wrap, and refrigerate for at least 1 hour.

Lightly grease the tart molds. Roll the pastry dough to 1/16” thick, into a rectangle of approximately 14” x 16”. Cut out 14 discs of 4” diameter each, and fit them into the tart molds. Refrigerate for 30 minutes.

Meanwhile, make the quiche filling: Melt 2/3 of the butter in a saucepan over medium heat, then sauté the sliced lemongrass and leeks with a dash of salt until soft, stirring regularly. Let cool. Melt 1.5 oz of butter in a saucepan over medium heat, then add 0.4 oz of water and asparagus, season with a dash salt, and cook for a couple of minutes until the asparagus is cooked but still crunchy. Let cool. In a bowl, mix the eggs, 1 egg yolk, heavy cream, curry, coriander, pepper and ¼ tsp of salt. Fold in the yogurt, the sautéd lemongrass and leek, and crumbled feta

Pour the quiche mixture into the molds, arrange 2 asparagus tips on top of each of them, and sprinkle with the grated Gruyère. Bake in a 400º F / 205º C oven for about 30 minutes, until golden brown.

Serve warm with some chopped parsley sprinkled on top.

The post Swiss Thai Quiche. A recipe from IBM’s Watson #CognitiveCooking appeared first on Incogvino.

The Chef at The Griffin telling the story of what it’s like to get his instructions from a computer…

You may remember when Watson won Jeopardy.

From Wikipedia:

Watson is an artificially intelligent computer system capable of answering questions posed in natural language,[2] developed in IBM’s DeepQA project by a research team led by principal investigator David Ferrucci. Watson was named after IBM’s first CEO and industrialist Thomas J. Watson.[3][4] The computer system was specifically developed to answer questions on the quiz show Jeopardy![5] In 2011, Watson competed on Jeopardy! against former winners Brad Rutter and Ken Jennings.[3][6] Watson received the first place prize of $1 million.[7]

Watson had access to 200 million pages of structured and unstructured content consuming four terabytes of disk storage[8] including the full text of Wikipedia,[9] but was not connected to the Internet during the game.[10][11] For each clue, Watson’s three most probable responses were displayed on the television screen. Watson consistently outperformed its human opponents on the game’s signaling device, but had trouble responding to a few categories, notably those having short clues containing only a few words.

Enter Watson’s next trick. It can cook. Which led to a fascinating experience… First of all, pairing wine to a bunch of recipes generated by a computer must have been interesting in itself, but probably the most interesting part of “Dining with Watson” was how hard it was for the Chef (in this case from The Griffin in Illovo) to put together these dishes. Some of them went against every trained cooking bone in his body. And yet… they tasted great!

Watson’s recipe generation repertoire is generated off a couple of factors.

• Primary ingredient
• Primary style (eg. Indian, Spanish etc.)
• Should the recipe be completely unique

Unique recipes is where it gets a bit crazy. Combinations of ingredients and styles that never, ever would make sense in your head. But I guess that’s the fun bit. And how you end up with Swiss Thai Quiche… or Belgian Bacon Pudding.

I’m going to post each recipe separately, but definitely have a look at the wine pairing below. Then follow the links below and try your hand at some computer generated cooking. Let us know how it goes!

IBM Watson’s recipe for Swiss Thai Quiche

IBM Watson’s recipe for Deboned Loin of Lamb with Pinenut Crust

IBM Watson’s recipe for Belgian Bacon Pudding

The post IBM’s Watson can cook (quite well actually)… appeared first on Incogvino.